JP2016210194A - Print data generation device, and computer program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To evaluate change in printing image quality due to pixel position shift in a printing medium.SOLUTION: A control device for a printing execution portion comprises: a half tone processing portion which executes half tone processing by which multiple gradation values constituting image data, whose gradation numbers are more than the types of formation states of printing pixel, are converted into dot data constituted of printing pixel values representing the formation states of the printing pixel; an error matrix acquisition portion which acquires plural types of error matrixes, which include a first type of error matrix which is used when pixel position shift is generated by a first amount and a second type of error matrix which is used when pixel position shift is generated by a second amount different from the first amount; and an evaluated image data generation portion which executes the half tone processing using the plural types of error matrixes respectively, and generates evaluated image printing data representing a plurality of evaluated images corresponding to the plural types of error matrixes respectively.SELECTED DRAWING: Figure 25

Description

  The present invention relates to image processing for printing, and more particularly to halftone processing using an error matrix.

  An ink jet that prints an image by forming dots (print pixels) by ejecting ink from the print head onto the paper while performing main scanning (also referred to as a pass) that moves the print head in the main scanning direction with respect to the paper. A printing device of the type is known. In addition, by irradiating the photosensitive drum with laser light, an electrostatic latent image is formed for each main scanning line along the main scanning direction of the laser light, and the toner adhered to the electrostatic latent image is transferred to the printing medium. Laser printers are known. In these printing apparatuses, the image quality of the print image may be deteriorated due to the shift of the print pixel position (pixel position in the print medium). In order to suppress this deterioration in image quality, a technique for printing a plurality of evaluation images while changing the timing of ejecting ink during main scanning in an ink jet printing apparatus is known (for example, Patent Document 1). ).

JP 2009-56719 A

  According to this technique, it is possible to suppress deterioration in image quality of a print image by adjusting the timing of ejecting ink during main scanning based on the image quality evaluation results of these evaluation images.

  However, with the above technique, there are cases where an evaluation image that can sufficiently evaluate the image quality of a printed image cannot be created. For example, if the ejection timing cannot be adjusted with sufficient accuracy with respect to the printing conditions such as the main scanning speed and the printing resolution, an appropriate evaluation image may not be created. Such a problem is also a problem common to laser printers. For example, if adjustment of the timing of laser light irradiation cannot be performed with sufficient accuracy, an appropriate evaluation image may not be created. .

  A main advantage of the present invention is to provide a new technique for evaluating a change in print image quality caused by a shift in pixel position on a print medium in a printing apparatus.

  SUMMARY An advantage of some aspects of the invention is to solve at least a part of the problems described above, and the invention can be implemented as the following application examples.

Application Example 1 A printing control apparatus for a printing execution unit that forms printing pixels on a printing medium using a printing material,
Dot data that is a multi-gradation value that constitutes image data and has a number of gradations that is greater than the type of print pixel formation state, and that is composed of print pixel values that represent the print pixel formation state A halftone processing unit that executes a halftone process for converting to a halftone process, wherein the halftone process uses the multi-tone value of a pixel to be processed and a distribution error to use the print pixel value of the pixel to be processed. The distribution error is an error generated between the print pixel value in the determined pixel for which the print pixel value has been determined and the multi-gradation value in the determined pixel. The error is distributed to the processing target pixel using an error matrix, and the error matrix is a matrix that defines a distribution ratio of errors in peripheral pixels of the processing target pixel. And parts,
An error matrix acquisition unit that acquires a plurality of types of error matrices, wherein the plurality of types of error matrices include a first type of error matrix used when a pixel position shift occurs by a first amount, and the first type of error matrix. A second type of error matrix used when the pixel position shift occurs by a second amount different from the amount of the pixel, and the pixel position deviation includes the pixel position in the main scanning direction in the image data and the print medium. The error matrix acquisition unit, wherein the main scanning direction pixel position is a position along the main scanning direction of the peripheral pixels with respect to the processing target pixel; and
An evaluation image data generation unit that generates evaluation image print data representing a plurality of evaluation images corresponding to each of the plurality of types of error matrices by executing the halftone process using each of the plurality of types of error matrices. When,
A printing control apparatus.

  According to the above configuration, a plurality of evaluation images capable of evaluating a change in print image quality caused by a shift (pixel position shift) in the main scanning direction pixel position in the print medium with respect to a main scanning direction pixel position in the image data. Can be easily printed by the print execution unit.

  The present invention can be realized in various forms. For example, a printing apparatus including the print data generation apparatus and the print execution unit, a print data generation method, a function of these apparatuses, or the method described above. The present invention can be realized in the form of a computer program for realizing, a recording medium on which the computer program is recorded, and the like.

1 is a block diagram of a multifunction peripheral 600 as an embodiment of the present invention. FIG. 2 is a diagram illustrating a schematic configuration of a print execution unit 200 of a multifunction peripheral 600. FIG. 4 is a diagram for explaining landing positions on a sheet of ink ejected from a print head. 10 is a flowchart of print processing executed by the multifunction peripheral 600. The figure explaining the bidirectional | two-way shift | offset | difference at the time of bidirectional | two-way interlaced printing. Schematic of halftone processing. The flowchart of a halftone process. The figure which shows the example of the error matrix used when the bidirectional | two-way deviation has arisen. It is a figure explaining input value correction. It is a figure explaining the effect of 1st Example. It is a figure explaining the main scanning direction shift | offset | difference of a nozzle. It is a figure explaining the opposing direction shift | offset | difference of a nozzle. The figure explaining head horizontal inclination shift. The figure which shows the example of the error matrix for head inclination deviation. The figure explaining the head vertical inclination shift at the time of bidirectional non-interlaced printing. The figure explaining the composite shift | offset | difference of the bidirectional | two-way shift | offset | difference and head horizontal inclination shift | offset | difference at the time of bi-directional interlace printing. The figure which shows the example of the error matrix used when the composite pixel position shift has arisen. The figure explaining the composite shift | offset | difference of the bidirectional | two-way shift | offset | difference at the time of bidirectional | two-way interlaced printing, and a head perpendicular | vertical inclination shift. The expansion perspective view of the main scanning part 240 and the sub-scanning part 260 in 4th Example. The figure explaining the halftone process in 4th Example. The figure which shows the outline | summary of an error collection method. The figure which shows the example of the use error matrix for error collection methods. The figure which shows the example of the use error matrix for error collection methods. 6 is a flowchart of evaluation image printing processing. Schematic which shows the content of print image IM1 for evaluation. The figure explaining the use error matrix used for the production | generation of evaluation image data. 6 is a flowchart showing processing steps of print setting processing. Schematic which shows the content of print image IM2 for evaluation.

A. First embodiment:
A-1: Configuration of MFP 600 Next, an embodiment of the present invention will be described based on examples. FIG. 1 is a block diagram of an MFP 600 as an embodiment of the present invention. The multi-function device 600 includes a control device 100, a print execution unit 200, and a scanner unit 400.

  The control device 100 is a computer that controls the operation of the MFP 600. The control device 100 communicates with a CPU 110, a volatile memory 120 such as a DRAM, a nonvolatile memory 130 such as an EEPROM, an operation unit 170 such as a touch panel, a display unit 180 such as a liquid crystal display, and an external device. And a communication unit 190 which is an interface of the above. The communication unit 190 may be, for example, a so-called USB interface or an interface compliant with IEEE802.3.

  The nonvolatile memory 130 stores a program 132, error matrix data 136 that defines an error matrix, and correction coefficient data 138 that defines a positional deviation correction coefficient Pc described later. The CPU 110 implements various functions including the function of the print data generation unit M100 by executing the program 132. The print data generation unit M100 causes the print execution unit 200 to execute printing using image data representing a print target image. The target image data is, for example, image data supplied to the multi-function device 600 from an external device (for example, a computer or a USB flash memory).

  In the present embodiment, the print data generation unit M100 includes an image data acquisition unit M102, a multi-tone value correction unit M104, a halftone processing unit M106, and an error matrix acquisition unit M108. Furthermore, the print data generation unit M100 may include an evaluation image data generation unit M110, an image quality evaluation acquisition unit M112, and a print setting determination unit M114. Processing executed by each of these functional units will be described later.

The scanner unit 400 receives transmitted light or reflected light from an object (original) by a photoelectric conversion element (for example, a CCD (Charge Coupled Device)), and generates image data (scan data) representing the object. .

  The print execution unit 200 forms ink dots on the print medium by ejecting ink toward the print medium. The print execution unit 200 includes a control circuit 210, a print head 250, a main scanning unit 240, and a sub scanning unit 260.

The print execution unit 200 will be further described with reference to FIG. 2 in addition to FIG. FIG. 2 is a diagram illustrating a schematic configuration of the print execution unit 200 of the multi-function device 600. FIG. 2A shows an outline of the overall configuration of the print execution unit 200, and FIG. 2B shows the configuration of the nozzle formation surface of the print head 250. In FIG. 2, three directions Dx, Dy, Dz are shown. The two directions Dx and Dy are horizontal directions, respectively, and the Dz direction is a vertically upward direction. The Dy direction and the Dx direction are orthogonal to each other. Hereinafter, the Dx direction is also referred to as “+ Dx direction”, and the opposite direction of the Dx direction is also referred to as “−Dx direction”. The same applies to the + Dy direction, the -Dy direction, the + Dz direction, and the -Dz direction.

  The sub-scanning unit 260 performs sub-scanning for transporting the print medium 300 (for example, A3 or A4 size paper) in the −Dx direction. As a result, the print head 250 moves in the + Dx direction with respect to the print medium 300 (hereinafter, the + Dx direction is also referred to as “sub-scanning direction”).

  Specifically, the sub-scanning unit 260 includes a transport motor 263, a first roller 261, a second roller 262, and a platen 265. The platen 265 supports the print medium 300 horizontally from the lower surface. The two rollers 261 and 262 are arranged at positions facing the upper surface of the platen 265. The first roller 261 faces the + Dx direction side portion of the platen 265 when viewed from the print head 250, and the second roller 262 faces the −Dx direction side portion of the platen 265 when viewed from the print head 250. Each of the rollers 261 and 262 is a roller parallel to the Dy direction, and is rotationally driven by the transport motor 263. The print medium 300 is sandwiched between the rollers 261 and 262 and the platen 265 and is conveyed in the −Dx direction by the rotating rollers 261 and 262.

  The main scanning unit 240 performs main scanning that causes the print head 250 to reciprocate in a direction (+ Dy direction or −Dy direction) orthogonal to the sub-scanning direction. Hereinafter, the + Dy direction and the −Dy direction are also referred to as “main scanning direction”, and among the main scanning directions, the + Dy direction is also referred to as “forward direction”, and the −Dy direction is also referred to as “return direction”.

  Specifically, the main scanning unit 240 includes a moving motor 242 and a support shaft 244. The support shaft 244 is disposed between the first roller 261 and the second roller 262 and extends in parallel with the main scanning direction. The support shaft 244 supports the print head 250 slidably along the support shaft 244. The moving motor 242 is connected to the print head 250 by a belt (not shown) and supplies power for main scanning.

  The print head 250 includes a plurality of nozzle arrays NC, NM, NY, and NK on the surface facing the print medium 300 (FIG. 2B). The plurality of nozzle arrays NC, NM, NY, and NK eject inks of four colors (cyan (C), magenta (M), yellow (Y), and black (K)) used for printing. Each of the nozzles 250n has a plurality of (for example, 200) nozzles 250n that discharge ink of the same color to form dots on the print medium 300. Each nozzle 250n is driven by the nozzle 250n. Piezo elements (not shown) are provided as drive elements for ejecting the nozzles A plurality of nozzles 250n included in one nozzle row are arranged at a nozzle pitch N in the sub-scanning direction. 2B, the nozzles do not have to be arranged in a straight line, but may be arranged in a zigzag, for example, in order to avoid complication of the drawing, the print heads shown in FIG. In the figure of 250 , Few nozzles 250n for discharging one kind of ink (e.g., four) using simplified diagram with a.

The control circuit 210 includes a main scanning control unit 212 that controls the main scanning unit 240, a head driving unit 214 that drives the print head 250 to eject ink from the nozzles 250n, and a sub scanning control unit that controls the sub scanning unit 260. 216. The control circuit 210 controls the main scanning unit 240, the print head 250, and the sub-scanning unit 260 according to the print data supplied from the control device 100, and executes printing. Specifically, the control circuit 210 performs image printing by alternately repeating unit printing and unit sub-scanning. Unit printing is printing performed by driving the print head 250 and ejecting ink while performing main scanning while the printing medium 300 is stopped. One main scan corresponding to one unit printing is also called a pass. As the unit printing described above, the control circuit 210 performs forward printing for performing printing while performing main scanning in the forward direction (also referred to as forward path) and returning for performing printing while performing main scanning in the backward direction (also referred to as backward path). Printing. Unit sub-scanning is to execute sub-scanning by a predetermined unit feed amount.

  FIG. 3 is a diagram illustrating the landing positions on the sheet of the inks I1 and I2 ejected from the print head 250. FIG. The first ink I1 indicates ink ejected from the print head 250 during the forward pass, and the second ink I2 indicates ink ejected from the print head 250 during the backward pass.

  3A shows a case where both the first ink I1 and the second ink I2 ideally land on the target position Py1 of the print medium 300. FIG. Here, when the ink ejection timing deviates from an ideal timing in at least one of the forward pass and the backward pass, the landing position (dot formation position) of the first ink I1 and the landing position of the second ink I2 are relatively Shift. In this way, the positional deviation that occurs between the dot D1 printed in the forward pass and the dot D2 printed in the backward pass is referred to as bidirectional deviation hereinafter.

  FIG. 3B shows a case where the ink ejection timing is delayed in both the forward pass and the return pass as compared with the case of FIG. In this case, the landing position of the first ink I1 is shifted in the forward direction from the target position Py1. The landing position of the second ink I2 deviates in the backward direction from the target position Py1. Therefore, in this example, when viewed from the dot D1 (pixel) of the forward raster line that is the raster line (main scanning line) printed in the forward path, the dot D2 of the backward raster line that is the raster line printed in the backward path. Will deviate in the return direction. Then, when viewed from the dot D2 of the backward raster line, the dot D1 of the forward raster line is shifted in the forward direction. Hereinafter, this bidirectional displacement is referred to as a positive bidirectional displacement.

  Contrary to the example of FIG. 3B, in contrast to the case of FIG. 3A, when the ink ejection timing is earlier in both the forward pass and the return pass, the dot D1 of the forward raster line. , The dot D2 on the return raster line is shifted in the forward direction. Then, when viewed from the dot D1 of the backward raster line, the dot D2 of the forward raster line is shifted in the backward direction. Hereinafter, this bidirectional displacement is referred to as a negative bidirectional displacement.

  Note that a gap GP in FIG. 3 indicates a gap between the nozzle surface np and the print medium 300. Here, the nozzle surface np is a plane including the positions of the plurality of nozzles 250n, and is a plane including a position through which the moving nozzle 250n can pass. The gap GP is a distance between the nozzle surface np and the print medium 300.

  In some cases, the image quality of a printed image is deteriorated due to the above-described dot position shift (also referred to as pixel position shift). Generally, in order to print an image, a plurality of dots separated from each other are formed. For example, due to pixel position deviation, some dots may approach other dots, and some dots may be connected to other dots. As a result, the printed image may appear rough compared to the case where there is no pixel position shift (the grain (dot) pattern representing the printed image may appear rough). That is, there is a case where the graininess is deteriorated as the image quality deterioration due to the pixel position shift. In order to suppress degradation in image quality due to pixel position deviation, the print data generation unit M100 of the multifunction peripheral 600 performs halftone processing using a specific error matrix as will be described later.

A-2: Overview of print processing:
FIG. 4 is a flowchart of a printing process executed by the multi-function device 600 (FIG. 1). The print data generation unit M100 starts print processing in response to a user instruction (for example, an operation of the operation unit 170 by a user or a command from a computer (not shown) connected to the communication unit 190). The image data acquisition unit M102 of the print data generation unit M100 acquires target image data to be printed from an external device connected to the communication unit 190.

  In step S200, the image data acquisition unit M102 converts the target image data into bitmap data BD (rasterization processing). The pixel data constituting the bitmap data BD is, for example, a pixel value with gradation values (for example, 256 gradations of 0 to 255) of three color components of red (R), green (G), and blue (B). This is RGB pixel data representing a color. In this embodiment, the resolution of the bitmap data BD is the same as the printing resolution (dot recording resolution).

  In step S210, the image data acquisition unit M102 converts the RGB pixel data constituting the bitmap data BD into gradation values (component values) of four color components (C, M, Y, K) corresponding to the ink colors. Are converted into CMYK pixel data representing the color of the pixel (color conversion processing). The color conversion process is performed using a lookup table that associates RGB pixel data with CMYK pixel data. The number of gradations of each component value of the CMYK pixel data is larger than the number of types of dot formation states (in this embodiment, four kinds as described later) (for example, 256 gradations from 0 to 255).

  In step S220, halftone processing is executed. The halftone process is a process of converting CMYK pixel data constituting the bitmap data BD into dot data (print pixel value) representing a dot (print pixel) formation state for each pixel. Halftone processing is executed using an error diffusion method using an error matrix. Details of step S220 will be described later.

  In step S230, the print data generation unit M100 generates print data from the dot data. The print data is data expressed in a data format that can be interpreted by the print execution unit 200. The print data generation unit M100 rearranges the dot data in the order used for printing according to a printing method such as bidirectional interlaced printing described later, and adds various printer control codes and data identification codes. Generate print data.

  In step S240, the print data generation unit M100 supplies the print data to the print execution unit 200. The print execution unit 200 prints an image according to the received print data.

A-3: Halftone processing:
In the halftone processing in the present embodiment, the error matrix to be used may differ depending on the printing method. In this embodiment, a case where the printing method is bidirectional interlaced printing will be described as an example. FIG. 5 is a diagram for explaining bidirectional displacement during bidirectional interlaced printing. In the center of FIG. 5, the bitmap data BD to be processed is conceptually shown. The cell PX represents a pixel constituting the bitmap data BD. The left side of FIG. 5 shows the position of the print head 250 in the sub-scanning direction in each pass when printing the bitmap data BD. The four nozzles 250n of the print head 250 are also called nozzles 1, 2, 3, and 4 in order from the upper side in FIG. “M” in the path code Pm in FIG. 5 represents the order in which the paths are executed. In FIG. 5, following the path code Pm, (F) is added when the path is a forward path, and (R) is appended when the path is a return path.

As can be seen from FIG. 5, bidirectional interlaced printing in this embodiment is two-pass printing in which printing is performed with a dot pitch that is half the nozzle pitch N. In this bidirectional interlaced printing, odd-numbered paths are forward paths and even-numbered paths are return paths. Further, unit sub-scanning corresponding to the dot pitch is performed between the forward pass and the next return pass, and the head length (in the example of FIG. 5, the nozzle pitch is set) between the return pass and the next forward pass. Unit sub-scan for 4 times N) is performed. As shown in FIG. 5, an odd-numbered raster line (from the upper end of the bitmap data BD)
Odd-numbered lines) are printed in the forward path, and even-numbered raster lines (even-numbered lines) are printed in the return path.

  In the present embodiment, it is assumed that a bidirectional displacement in the positive direction of the displacement amount s occurs. The right side of FIG. 5 illustrates the positional relationship between the dot D1 (single circle) printed in the forward path and the dot D2 (double circle) printed in the backward path as an explanation of the bidirectional displacement. Yes. These dots D1 and D2 are ideally dots that should be aligned along the sub-scanning direction (when there is no bi-directional deviation). The number in each dot indicates the number of the nozzle that prints that dot (left side in FIG. 5). For example, a double circle in which “2” is described indicates a dot printed by the nozzle 2 in the return pass.

  Further, on the right side of FIG. 5, as an explanation of the bidirectional displacement, a box ZB representing an aspect of the relative pixel position displacement of the raster line is illustrated. This box ZB represents the pixel position shift of the first adjacent raster line with respect to the reference raster line and the pixel position shift of the second adjacent raster line with respect to the reference raster line. The first adjacent raster line is a raster line adjacent to the reference raster line in the sub-scanning direction. The second adjacent raster line is a raster line positioned in the sub-scanning direction by two lines from the reference raster line (raster line adjacent to the first adjacent raster line in the sub-scanning direction). Box ZB includes three cells. The first cell in which the “*” mark is described represents the position of the reference raster line. In the second grid adjacent in the sub-scanning direction of the first grid, the pixel position shift of the first adjacent raster line is described. In the third grid adjacent in the sub-scanning direction of the second grid, the pixel position shift of the second adjacent raster line is described. As understood from the contents of the box ZB, the pixel position shift of the second adjacent raster line with respect to each raster line is “0”. Then, it is understood that the pixel position shift of the return raster line with respect to the forward raster line is “−s”, and the pixel position shift of the forward raster line with respect to the backward raster line is “+ s”. A pixel position shift of “+ s” indicates that the shift amount is s in the forward direction (+ Dy direction), and a pixel position shift of “−s” is a shift amount s in the return path direction (−Dy direction). Represents that

  FIG. 6 is a schematic diagram of halftone processing. FIG. 7 is a flowchart of halftone processing. Halftone processing (FIG. 7) is executed for each ink color and for each pixel. A broken line PL in FIG. 5 indicates the order of pixels to be processed in the halftone process for one ink color. In this embodiment, the pixels PX on the odd-numbered lines are sequentially processed in the forward direction (+ Dy direction), and the pixels PX on the even-numbered lines are sequentially processed in the backward direction (−Dy direction). . In the case of this processing order, an error matrix for forward direction processing is used for pixels PX on odd-numbered lines, and an error matrix for backward direction processing is used for pixels PX on even-numbered lines. The error matrix for forward direction processing (for example, the matrix MAa in FIG. 5) and the error matrix for backward direction processing (for example, the matrix MAb in FIG. 5) are generally symmetric in the main scanning direction. In the example of FIG. 5, although coincidentally coincident, the processing direction of halftone processing of a certain raster line and the direction of the pass when printing the raster line are irrelevant. In all embodiments, the processing direction (processing order) of the halftone processing is the same regardless of the printing method (bidirectional printing, unidirectional printing, etc.).

By this halftone process, dot data representing the dot formation state is generated. In this embodiment, the dot formation state is determined from the following four states in which the ink amount (that is, the dot size (density expressed by the dot)) is different.
A) Large dot B) Medium dot (Medium dot density <Large dot density)
C) Small dot (small dot density <medium dot density)
D) No dot Therefore, the number of gradations of the dot data is 4.

  In step S502, the error matrix acquisition unit M108 acquires a misregistration correction coefficient Pc (also simply referred to as a correction coefficient Pc) for bidirectional deviation recorded in the correction coefficient data 138. For example, the correction coefficient Pc corresponds to the magnitude and direction of bidirectional shift in the print execution unit 200 described above. For example, the correction coefficient Pc is expressed in units of the dot pitch in the main scanning direction. If the correction coefficient Pc accurately corresponds to the bidirectional shift described in FIG. 5, the correction coefficient Pc = + s. The correction coefficient Pc is determined, for example, by specifying the value of “s” described in FIG. 5 by measuring the bi-directional deviation actually occurring in the printed test image. Alternatively, as described later, it may be determined based on the result of image quality evaluation (for example, graininess evaluation) for the printed evaluation image.

  In step S504, the error matrix acquisition unit M108 acquires an error matrix to be used. FIG. 8 is a diagram illustrating an example of an error matrix (also referred to as an error matrix for bidirectional displacement) used when bidirectional displacement occurs. FIG. 8A shows standard error matrices MAa and MAb. The error matrix whose code ends with “a” indicates the error matrix for forward direction processing described above, and the error matrix whose code ends with “b” indicates the above-described error matrix for backward direction processing. Hereinafter, when the forward direction processing and the backward direction processing are not distinguished, “a” or “b” at the end of the reference numerals is appropriately omitted. The standard error matrix MA is an error matrix designed so as to realize a suitable image quality when no pixel position deviation occurs.

  The error matrix defines an error distribution ratio for peripheral pixels located within a specific range when viewed from the processing target pixel. The error matrix of this embodiment includes line matrices LM1 to LM3 for three rows, like the standard error matrix MA. The first line matrix LM1 defines the error distribution ratio for the peripheral pixels located on the target raster line where the processing target pixel is located. The second line matrix LM2 defines the error distribution ratio for the peripheral pixels located on the first adjacent raster line of the target raster line. The third line matrix LM3 defines the error distribution ratio for the peripheral pixels located on the second adjacent raster line of the target raster line.

  The error matrix acquisition unit M108 creates a use error matrix by shifting the shift target line matrix of the standard error matrix MA in the main scanning direction using the correction coefficient Pc. The shift target line matrix is a line matrix corresponding to a raster line that is a raster line different from the target raster line and in which peripheral pixels in which a pixel position shift occurs with respect to the processing target pixel. Specifically, the error matrix acquisition unit M108 shifts the shift target matrix so as to cancel the relative shift shown in the box ZB on the right side of FIG. The shift target matrix is a line matrix corresponding to a raster line having a relative shift with respect to the processing target pixel in the line matrices LM2 and LM3. To cancel the relative shift is to shift the shift target line matrix by the same amount as the relative shift in the opposite direction to the relative shift. In the case of the bi-directional shift of the present embodiment, the second line matrix LM2 is a shift target matrix, as can be seen from the box ZB on the right side of FIG.

The shift direction in which the shift target line matrix is shifted is opposite to the direction of pixel position deviation. As described above, in the case of a bidirectional shift in the positive direction (when the value of “s” in the box ZB on the right side of FIG. 5 is positive), the pixels of the return raster line are viewed from the pixels of the forward raster line. Shifts in the return direction (−Dy direction). Therefore, the use error matrix when the forward raster line (odd row line in this embodiment) is a processing target pixel is a matrix in which the second line matrix LM2 is shifted in the forward direction (+ Dy direction).
In the case of a bidirectional bi-directional shift, the pixels of the forward raster line are shifted in the forward direction (+ Dy direction) when viewed from the pixels of the backward raster line. Therefore, the use error matrix when the return raster line (even line in this embodiment) is a processing target pixel is a matrix in which the second line matrix LM2 is shifted in the return direction (−Dy direction).

  The shift amount by which the shift target line matrix is shifted is preferably the same as the pixel position shift amount. In this embodiment, the error matrix acquisition unit M108 shifts the shift target line matrix by the amount “s” (FIG. 5) of the pixel position deviation.

  FIG. 8B shows a use error matrix MB when the bidirectional deviation is +1 pixel, that is, when s = + 1. It can be seen that the second line matrix LM2 of the use error matrix MBa is shifted by one pixel in the forward direction as compared to the second line matrix LM2 (FIG. 8A) of the standard error matrix MAa. It can be seen that the second line matrix LM2 of the use error matrix MBb is shifted by one pixel in the backward direction compared to the second line matrix LM2 of the standard error matrix MAb (FIG. 8A).

  As shown in the examples of FIG. 8A (before the shift) and FIG. 8B (after the shift), the shift of the shift target line matrix when the shift amount is an integer value is the error of each pixel before the shift. This is performed by moving the distribution ratio to pixels located in the shift direction by the integer value (referred to as integer shift).

  In the case of bidirectional printing, a combination of two error matrices (also referred to as an error matrix set) in which the shift direction of the shift target line matrix is opposite to each other is used as the use error matrix. In the example of FIG. 8B, the right error matrix MBa and the left error matrix Mb are combined. One of the error matrix sets (for example, the error matrix MBa) is used when the raster line where the processing target pixel is located is printed in the forward pass, and the raster line corresponding to the shift target line matrix is printed in the backward pass. It is done. The other of the error matrix sets (for example, the error matrix MBb) is used when the raster line where the processing target pixel is located is printed in the backward pass, and the raster line corresponding to the shift target line matrix is printed in the forward pass. It is done.

  FIG. 8B shows a use error matrix MC when the bidirectional displacement is +0.5 pixels, that is, when s = + 0.5. In this example, the use error matrix MC is obtained by shifting the shift target matrix (the second line matrix LM2 of the standard error matrix MA) by 0.5 pixels.

  A method of shifting (referred to as a fractional shift) when the shift amount N is in the range of 0 <N <1 as in this example will be described. First, the error matrix acquisition unit M108 subtracts the multiplication distribution ratio obtained by multiplying the distribution ratio of each pixel defined in the shift target line matrix by the shift amount N from the distribution ratio of each pixel. Next, the error matrix acquisition unit M108 adds the multiplication distribution ratio to the distribution ratio of pixels adjacent in the shift direction of each pixel. By these two processes, the error matrix acquisition unit M108 shifts the shift target matrix by a decimal number.

Example of calculating the use error matrix MCa (FIG. 8C) by shifting the second line matrix LM2 of the standard error matrix MAa (FIG. 8A) by 0.5 pixels in the forward direction (+ Dy direction). Will be described. In this example, the shift target matrix before the shift (the second line matrix LM2 of the standard error matrix MAa) is expressed as LM = (0, 1, 2, 1, 0) (FIG. 8A). Here, the five components of the matrix LM are error distribution ratios of the peripheral pixels PX1, PX2, PX3, PX4, and PX5 shown in FIG. When the shift amount N = 0.5, the matrix K = N × LM = (0, 0.5, 1, 0.5, 0) representing the multiplication distribution ratio. A matrix KS = (0, 0, 0.5, 1, 0.5) obtained by moving the multiplication distribution ratio of each peripheral pixel to a pixel adjacent in the shift direction (+ Dy direction). Therefore, when the shift target matrix after the shift (the second line matrix LM2 of the use error matrix MCa) is LMs, LMs = LM−K + KS.
Therefore, LMs = (0, 1, 2, 1, 0) − (0, 0.5, 1, 0.5, 0) + (0, 0, 0.5, 1, 0.5) = (0 , 0.5, 1.5, 1.5, 0.5) (FIG. 8C).

  By using the same calculation, the use error matrix MCb can be calculated by shifting the second line matrix LM2 of the standard error matrix MAb by 0.5 pixels in the backward direction (−Dy direction). Since the error distribution ratio represents a ratio, even if each error distribution ratio is multiplied by the same number, it is substantially synonymous. Therefore, when the error distribution ratios of the upper use error matrices MCa and MCb in FIG. 8C are multiplied by 2, the same use error matrices MC2a and MC2b in which the error distribution ratios are expressed as integers are obtained (FIG. 8 ( C) Lower side).

  As can be understood from the above description, by performing the above-described fractional shift, the error matrix acquisition unit M108 can execute the shift of the shift target line matrix with an arbitrary accuracy smaller than one pixel. That is, the error matrix acquisition unit M108 can acquire a use error matrix corresponding to an arbitrary pixel position shift smaller than one pixel. As an example, FIG. 8D shows a use error matrix MD in the case where the bi-directional deviation is +1/16 pixel, that is, the correction coefficient Pc = + 1/16.

  More generally, when the shift amount N = N1 + N2 (N1 is an integer, N2 is a value of 0 <N2 <1 (decimal component)), the error matrix acquisition unit M108 performs N1 with respect to the shift target line matrix. Only the integer shift is performed, and the shift is performed by the fractional number by N2 with respect to the shift target line matrix after the integer shift.

  Returning to FIG. 7, the description will be continued. When the use error matrix is acquired, in step S506, the multi-tone value correcting unit M104 corrects the input value Vin1 (the component value of the ink color to be processed among the CMYK pixel data of the processing target pixel), and The correction input value Vin2 is acquired. The multi-tone value correction unit M104 corrects the input value Vin1 so that the raster line is shifted in the direction opposite to the direction of pixel position deviation. The multi-tone value correction unit M104 corrects the input value Vin1 using the correction coefficient Pc.

  FIG. 9 is a diagram for explaining the input value correction. FIG. 9 shows input values for each pixel PX constituting the bitmap data BD. FIG. 9A shows an input value before correction. FIG. 9B shows an example of an input value after correction in the case of bidirectional deviation + 1 pixel, that is, in the case of s = + 1. When s = + 1, in the print image, the backward raster line (even line in this embodiment) is shifted by one pixel in the + Dy direction with respect to the forward raster line (odd line in this embodiment). Therefore, the multi-tone value correction unit M104 corrects the input value Vin1 so that the backward raster line is shifted by one pixel in the −Dy direction with respect to the forward raster line. In the example of FIG. 9B, the multi-gradation value correcting unit M104 shifts the pixels of the forward raster line by +0.5 pixels in the + Dy direction, and the pixels of the backward raster line by 0.5 pixels. The input value Vin1 is corrected so as to shift in the Dy direction. The multi-gradation value correction unit M104 corrects the input value Vin1 so that the pixels of the forward raster line are not subjected to the input value correction, and the pixels of the backward raster line are shifted by one pixel in the −Dy direction. May be.

When the correction amount Q (shift amount) is an integer, the multi-gradation value correction unit M104 inputs pixels positioned in the opposite direction of the shift direction by the correction amount Q (integer) as viewed from the processing target pixel. Value Vi
Let n1 be the correction input value Vin2 of the pixel to be processed. When the correction amount Q is a decimal number (0 <Q <1), the multi-tone value correction unit M104 performs input value correction using the following equation (1).
Vin2 = (1-Q) × Vin1 + Q × Vne (1)
Here, Vne is an input value of a pixel adjacent in the direction opposite to the shift direction when viewed from the processing target pixel.

  FIG. 9C shows an example of an input value after correction in the case of bidirectional deviation +0.5 pixels, that is, in the case of s = + 0.5. In this example, the multi-gradation value correcting unit M104 shifts the forward raster line pixels in the + Dy direction by 0.25 pixels, and shifts the backward raster line pixels in the −Dy direction by 0.25 pixels. Thus, the input value Vin1 is corrected.

  Returning to FIG. 7, the description will be continued. In step S508, the halftone processing unit M106 acquires the corrected input value Va by adding the distribution error value Et to the corrected input value Vin2. The distribution error value Et is acquired from the error buffer EB. The distribution error value Et will be described later.

  In step S510, the halftone processing unit M106, based on the magnitude relationship between the corrected input value Va and the three threshold values Th1 to Th3, dot data Dout (print pixel value representing a dot formation state) of the processing target pixel. To decide. The threshold values Th1 to Th3 are 1, 85, and 170, respectively. The halftone processing unit M106 determines to form a small dot when the corrected input value Va is greater than or equal to the first threshold Th1 and less than the second threshold Th2. The halftone processing unit M106 determines to form a medium dot when the corrected input value Va is equal to or greater than the second threshold Th2 and less than the third threshold Th3. The halftone processing unit M106 determines to form a large dot when the corrected input value Va is equal to or greater than the third threshold Th3.

In step S512, the halftone processing unit M106 converts the dot data Dout into a relative dot value Dr. The relative dot value Dr is a gradation value associated with four values (representing the formation state of four types of dots) that the dot data Dout can take. The relative dot value Dr is a value that represents the density expressed by the formation state of the four types of dots with a density of 256 gradations. In this embodiment, the relative dot value Dr is set as follows.
A) Large dot: Relative dot value Dr = 255
B) Medium dot: Relative dot value Dr = 170
C) Small dot: Relative dot value Dr = 85
D) No dot: relative dot value Dr = 0
The relative dot value Dr is incorporated in advance in the program of the halftone processing unit M106 as the relative value table DT.

In step S514, the halftone processing unit M106 distributes the target error value Ea to the peripheral pixels according to the use error matrix. Specifically, the halftone processing unit M106 calculates the target error value Ea with the following equation.
Target error value Ea = corrected input value Va−relative dot value Dr
The target error value Ea can be said to be an error generated between the dot data in the processing target pixel (dot data converted into the relative dot value Dr) and the input value (corrected input value Va) in the processing target pixel. The halftone processing unit M106 distributes the target error value Ea to surrounding pixels at a distribution ratio defined in the above-described use error matrix. In the error buffer EB, a cumulative addition value of the target error value Ea distributed in step S514 is recorded for each unprocessed pixel that is not a target of halftone processing. The distribution error value Et acquired in step S508 described above is a cumulative addition value of the target error value Ea recorded in the error buffer EB for the processing target pixel. That is, the distribution error value Et is distributed to the processing target pixels using the use error matrix among the target error values Ea in the processed pixels (determined pixels for which dot data has been determined) that are the processing target of the halftone process. Is a cumulative addition value of the error.

  By the halftone process described above, bitmap data composed of dot data of print pixels is generated for each ink color.

  According to the first embodiment described above, the halftone processing unit M106 executes halftone processing using the use error matrix. The use error matrix is adjusted according to the pixel position deviation on the print medium 300 when printing on the print medium 300 with respect to the pixel position in the main scanning direction in the bitmap data BD. As a result, even when the pixel position shift occurs, it is possible to suppress a decrease in print image quality due to the pixel position shift.

  Specifically, the second line matrix LM2 of the use error matrix is a matrix obtained by shifting the second line matrix LM2 of the standard error matrix in the direction opposite to the pixel position deviation. As a result, it is possible to suppress deterioration in image quality due to a pixel position shift between the processing target pixel and a pixel of a raster line different from the raster line where the processing target pixel is located.

  Since the shift amount of the second line matrix LM2 is the same as the amount of pixel position deviation, it is possible to effectively prevent the print image quality from being lowered according to the amount of pixel position deviation.

  When the amount of pixel position deviation is N pixels (0 <N <1), the second line matrix LM2 is shifted by the above-described decimal shift method, so that a pixel position deviation smaller than one pixel occurs. Even in this case, it is possible to easily suppress deterioration in image quality due to pixel position shift. In particular, if the print execution unit 200 cannot adjust the ink ejection timing with sufficient accuracy, for example, even if the print execution unit 200 does not have the ability to control with sufficient accuracy due to cost constraints, According to the present embodiment, it is possible to easily suppress deterioration in image quality due to pixel position deviation. In other words, by using the halftone process in the present embodiment, it is possible to reduce the cost of the print execution unit 200 without increasing the image quality deterioration due to the pixel position shift.

  The shift direction of the second line matrix LM2 includes an error matrix for pixels of the forward raster line (for example, a use error matrix MBa (FIG. 8)) and an error matrix for pixels of the backward raster line (for example, a use error matrix). MBb (FIG. 8)). As a result, it is possible to suppress a decrease in print image quality due to a positional shift that occurs between forward printing and backward printing in bidirectional printing.

  FIG. 10 is a diagram for explaining the effect of this embodiment. FIG. 10 shows an example where the bidirectional displacement is +1 pixel. FIG. 10A shows the pixel position in the image data (bitmap data BD), in other words, the pixel position when there is no pixel position shift (bidirectional shift in this embodiment). FIG. 10B shows a pixel position of a pixel position (a pixel position having a pixel position shift) in the print image PG on the print medium 300. In this embodiment, the shift target line matrix (second line) of the standard error matrix MA is applied to the multi-gradation value (gradation value of CMYK pixel data) at the pixel position (ideal pixel position) on the image data. A use error matrix MB (FIG. 8B) obtained by shifting the matrix LM2) in advance in consideration of the pixel position deviation is applied.

Then, the position on the image to which the target error value Ea is distributed becomes a position shifted in the opposite direction of the pixel position shift in the image data (no pixel position shift). As a result, when viewed in the print image PG (with pixel position deviation), the target error value Ea can be distributed to the position where the target error value Ea should be distributed. As shown in FIG. 10B, when viewed in the print image PG (with pixel position deviation), the distribution destination of the target error value Ea by the use error matrix MB is distribution by the standard error matrix MA when there is no pixel position deviation. You can see that it is the same as before.

  The same applies to the case where the shift target line matrix is shifted with a finer accuracy than one pixel in accordance with the pixel position deviation. FIG. 10C shows an example in which the shift amount is 0.5 pixels. Even in this case, when viewed in the printed image PG (with pixel position deviation), it is understood that the distribution destination of the target error value Ea based on the use error matrix MCa is close to the distribution destination based on the standard error matrix MAa when there is no pixel position deviation. . Specifically, the distribution ratios at the positions of the three black circles in FIG. These values are equal to the distribution ratio defined at the positions of the three black circles by the standard error matrix MAa when there is no pixel position deviation.

  As a result, particularly in an image region having a uniform density, the dot arrangement state in the print image PG (with pixel position deviation) is changed to the dot arrangement state realized by the standard error matrix MA when there is no image position deviation. You can get closer. As a result, it is possible to suppress deterioration in image quality due to pixel position shift, particularly in a region where the density is uniform. For example, since the deterioration of graininess is particularly noticeable in a region where the density is uniform, it is considered that the present embodiment has a great significance for improving the deterioration of graininess.

  Furthermore, in this embodiment, prior to the halftone process, the multi-tone value correction unit M104 corrects the input value Vin1 according to the pixel position deviation. As a result, it is possible to suppress deterioration in image quality (edge shading, etc.) at the edge portion of the printed image due to pixel position deviation. The halftone process using the above-described use error matrix has a lower image quality improvement effect at the edge portion of the printed image than an image quality improvement effect in a region having a uniform density. In the present embodiment, among the image quality degradation caused by the pixel position shift, the image quality degradation at the edge portion is corrected by correcting the input gradation value Vin, and the image quality degradation in the region where the density is relatively uniform is determined by using the above-described use error matrix. Each halftone process can be effectively improved.

  In this embodiment, the input value Vin1 is corrected regardless of whether or not it is an edge portion. However, the edge portion is detected by image processing, and the input value Vin1 is corrected only for pixels corresponding to the edge. May be performed. It is preferable to correct the input value Vin1 according to the pixel position shift at least for pixels in the edge portion of the image to be printed. However, since the correction method of the input value Vin1 in the present embodiment does not change the input value Vin1 in the region where the density is uniform, the input value Vin1 is substantially corrected only for the pixels corresponding to the edge. It can be said that it is going.

B. Second embodiment:
B-1: Pixel Misalignment In the second embodiment, halftone processing according to a pixel misalignment caused by a misalignment (nozzle misalignment) of the nozzle 250n in the print head 250 will be described. The configuration of the multi-function device 600 is the same as that of the multi-function device 600 (FIG. 1) of the first embodiment. The nozzle deviation considered in this embodiment is a main scanning direction deviation and a counter direction deviation.

FIG. 11 is a diagram for explaining the displacement of the nozzle in the main scanning direction. The displacement of the nozzle in the main scanning direction is that the position of the second nozzle is displaced from the designed position in the main scanning direction with respect to the position of the first nozzle. The first nozzle and the second nozzle are two of the plurality of nozzles 250n included in the nozzle row (FIG. 2B) for discharging the same color. As shown in FIG. 11, for example, the print head 250 is displaced in the main scanning direction (+
This occurs when tilted from a design position (horizontal tilt) along a plane parallel to the Dy direction and -Dy direction) and the sub-scanning direction (+ Dx direction). In other words, the displacement of the nozzle in the main scanning direction occurs, for example, when the print head 250 rotates from the designed position with the Dz direction as the rotation axis.

  In FIG. 11, four nozzles 250n arranged at equal intervals N in the sub-scanning direction (+ Dx direction) are nozzles 1, 2, 3, and 4 from the upstream side in the sub-scanning direction. The displacement of the nozzles in the main scanning direction due to the horizontal inclination of the print head 250 becomes larger as the nozzles are further away from the nozzles 1 when the nozzles 1 are used as a reference. For this reason, a pixel position shift (head horizontal tilt shift) caused by the horizontal tilt of the print head 250 is caused by a nozzle that is far from the nozzle 1 when the raster line pixel printed by the nozzle 1 is used as a reference. The raster line pixels to be printed become larger.

  In FIG. 11, dots D11 to D14 formed by ink ejected from the nozzles 1 to 4 in the forward pass, and dots D21 to D24 respectively formed from ink ejected from the nozzles 1 to 4 in the return pass. It is shown in the figure. As can be seen from FIG. 11, in the case of head horizontal tilt deviation, if the deviation amount of the dot D12 with respect to the dot D11 is k, the deviation amount of the dot 13 with respect to the dot D11 is 2k, and the deviation amount of the dot D14 with respect to the dot D11 is 3k. It becomes. Similarly, if the deviation amount of the dot D22 with respect to the dot D21 is k, the deviation amount of the dot 23 with respect to the dot D21 is 2k, and the deviation amount of the dot D24 with respect to the dot D21 is 3k. In the case of the head horizontal tilt shift, the pixel position shift direction when printing in the forward pass and the pixel position shift direction when printing in the return pass are the same direction.

  FIG. 12 is a diagram for explaining the displacement of the nozzles in the facing direction. The displacement of the nozzle in the facing direction is that the position of the second nozzle is displaced in the facing direction (Dz direction in FIG. 12) from the designed position with respect to the position of the first nozzle. The facing direction is a direction in which the nozzle (the nozzle forming surface of the print head 250) and the print medium 300 face each other. The first nozzle and the second nozzle are two of the plurality of nozzles 250n included in the nozzle row (FIG. 2B) for discharging the same color. For example, as shown in FIG. 12A, the nozzle facing direction deviation may occur when the print head 250 is inclined from a design position along a vertical plane including the sub-scanning direction (+ Dx direction) (vertical inclination). ). In other words, the nozzle facing direction deviation occurs, for example, when the print head 250 rotates from the designed position with the main scanning direction (+ Dy, -Dy direction) as the rotation axis.

  Due to the displacement of the nozzles in the opposite direction, a difference occurs in the gap between the nozzles and the print medium. For example, there is a difference between the gap GP1 between the nozzle 1 and the print medium 300 and the gap GP2 between the nozzle 2 and the print medium 300 (FIG. 12B). Due to this gap difference, there is a pixel position shift in the main scanning direction between the dots D11 and D21 formed by the ink discharged from the nozzle 1 and the dots D12 and D22 formed by the ink discharged from the nozzle 2. This occurs (FIG. 12B).

  As shown in FIG. 12A, the nozzle facing direction deviation due to the vertical inclination of the print head 250 becomes larger as the nozzle is more distant from the nozzle 1 when the nozzle 1 is used as a reference. For this reason, the pixel position deviation (head vertical inclination deviation) due to the vertical inclination of the print head 250 is caused by the nozzles that are separated from the nozzles 1 when the raster line pixels printed by the nozzles 1 are used as a reference. The raster line pixels to be printed become larger.

In FIG. 12C, as in FIG. 11, the dots D11 to D14 formed by the ink ejected from the nozzles 1 to 4 in the forward pass, and the ink ejected from the nozzles 1 to 4 in the return pass, respectively. The dots D21 to D24 formed by are illustrated. As can be seen from FIG. 12C, in the case of the head vertical inclination deviation, if the deviation amount of the dot D12 with respect to the dot D11 is h, the deviation amount of the dot 13 with respect to the dot D11 is 2h, and the deviation of the dot D14 with respect to the dot D11. The amount is 3h. Similarly, if the deviation amount of the dot D22 with respect to the dot D21 is h, the deviation amount of the dot 23 with respect to the dot D21 is 2h, and the deviation amount of the dot D24 with respect to the dot D21 is 3h. In the case of head vertical tilt deviation, the pixel position deviation direction when printing in the forward pass and the pixel position deviation direction when printing in the backward pass are opposite to each other.

B-2: Halftone processing:
In the halftone process in the present embodiment, the error matrix to be used may differ depending on the printing method, as in the first embodiment. In this embodiment, a case where the printing method is bidirectional non-interlaced printing will be described as an example.

B-2-1: When the head horizontal tilt shift occurs:
FIG. 13 is a diagram for explaining head horizontal tilt deviation during bidirectional non-interlaced printing. The left side of FIG. 13 shows the position of the print head 250 in the sub-scanning direction in each pass when bitmap data (not shown) is printed, as in FIG.

  As can be seen from FIG. 13, bidirectional non-interlaced printing in this embodiment is one-pass printing in which printing is performed at the same dot pitch as the nozzle pitch N. In this bidirectional non-interlaced printing, odd-numbered paths are forward paths and even-numbered paths are return paths. In this bidirectional non-interlaced printing, odd-numbered paths are forward paths and even-numbered paths are return paths. Further, unit sub-scan for the head length (four times the nozzle pitch N in the example of FIG. 13) is performed between one pass and the next pass.

  In this embodiment, it is assumed that a head horizontal tilt shift in the positive direction with a shift amount k occurs. FIG. 13 illustrates the positional relationship between the dot D1 printed in the forward pass and the dot D2 printed in the return pass, as in FIG. FIG. 13 further shows a box ZB as in FIG. In FIG. 13, a box ZB having the raster line of the basic area as a reference raster line (a position marked with *) and a box ZB having the raster line of the path boundary area as a reference raster line (a position marked with *) The contents are different. As described above, the use error matrix is created by shifting the shift target line matrix of the standard error matrix MA so as to cancel the relative shift shown in the box ZB. Therefore, in the case of the head horizontal tilt deviation, the usage error matrix applied to the raster line pixels in the basic area is different from the usage error matrix applied to the raster line pixels in the path boundary area. I understand that.

  The path boundary area is an area for two raster lines located at the boundary between the area printed in the forward path and the area printed in the backward path. The basic area is an area excluding the path boundary area. In the halftone process in which the pixels of the raster line included in the basic region are processed, three raster lines corresponding to the error matrix (the target raster line, the first adjacent raster line of the target raster line, and the target raster line All of the second adjacent raster lines are printed in the same path (forward path or return path). In the halftone process that targets the raster line pixels included in the path boundary area, some of the three raster lines corresponding to the error matrix are printed in the forward pass, and the other part is in the return pass. Printed. In FIG. 13, the number of nozzles of the print head 250 is set to four in order to avoid the complexity of the figure, but the actual number of nozzles is much larger than four (for example, 200). Most areas of the printed image are basic areas.

The pixel position shift of the first adjacent raster line with respect to each raster line in the basic region is “
k ”, and the pixel position shift of the second adjacent raster line is“ 2k ”. Thus, in the case of the head horizontal tilt deviation, the pixel position shift increases step by step according to the head tilt. In the case of the head horizontal tilt deviation, the position deviation mode (the contents of the box ZB) is the same in the area printed in the forward path and the area printed in the backward path. Therefore, the use error matrix in the area printed in the forward pass is the same as the use error matrix in the area printed in the return pass.

  FIG. 14 is a diagram illustrating an example of a use error matrix (also referred to as a head tilt shift error matrix) used when a head tilt shift occurs. FIG. 14A shows a standard error matrix MA. FIG. 14B shows a use error matrix ME when the head horizontal tilt deviation is (+1/16) pixels, that is, when k = + 1/16. The second line matrix LM2 of the use error matrices MEa and MEb is shifted by (1/16) pixels in the return path direction compared to the second line matrix LM2 of the standard error matrices MAa and MAb (FIG. 14A). I understand that you are doing. Then, the third line matrix LM3 of the use error matrices MEa and MEb is shifted by (2/16) pixels in the backward direction as compared with the third line matrix LM3 of the standard error matrices MAa and MAb. I understand. This use error matrix ME is applied to each raster line of the basic area in the area printed in the forward pass and the basic area in the area printed in the backward pass. As can be seen from the above description, in the case of head horizontal tilt deviation, the shift directions of the line matrices LM2 and LM3 are the same between the area printed in the forward pass and the area printed in the return pass. .

  The use error matrix applied to each raster line in the path boundary region is not shown, but the shift target line matrix of the standard error matrix MA so as to cancel the relative shift indicated by the box ZB in FIG. May be created by shifting. Alternatively, since the path boundary area is a smaller area than the basic area, the basic area use error matrix may be applied without creating a dedicated use error matrix.

B-2-2: When the head vertical tilt deviation occurs:
FIG. 15 is a diagram for explaining the head vertical tilt deviation during bidirectional non-interlaced printing. The left side of FIG. 15 shows the position of the print head 250 in the sub-scanning direction in each pass when bitmap data (not shown) is printed, as in FIG.

  In this embodiment, it is assumed that a head vertical tilt deviation in the positive direction with a deviation amount h occurs. FIG. 15 illustrates the positional relationship between the dot D1 printed in the forward pass and the dot D2 printed in the return pass, as in FIG. FIG. 15 further shows a box ZB as in FIG. In FIG. 15, the contents are different between a box ZB for a raster line in the basic area and a box ZB for a raster line in the path boundary area. Therefore, as in the case of the head horizontal tilt shift, in the case of the head vertical tilt shift, the usage error matrix applied to the raster line pixels in the basic area and the raster line pixels in the path boundary area are used. It can be seen that the applied error matrix is different.

As shown in FIG. 15, the pixel position shift of the first adjacent raster line with respect to each raster line in the basic region printed in the forward pass is “h”, and the pixel position shift of the second adjacent raster line is “2h”. It is. As described above, the mode of pixel position deviation in the area printed in the forward pass is the same as the mode of pixel position deviation in the case of head horizontal tilt deviation. On the other hand, the pixel position shift of the first adjacent raster line with respect to each raster line in the basic region printed in the return pass is “−h”, and the pixel position shift of the second adjacent raster line is “−2h”.
As described above, the direction of pixel position deviation in the area printed in the forward path is opposite to the direction of pixel position deviation in the case of head horizontal tilt deviation. Therefore, in the case of the head vertical inclination deviation, the use error matrix in the area printed in the forward pass is different from the use error matrix in the area printed in the return pass.

  FIG. 14C shows a use error matrix MF when the head vertical tilt deviation is (+1/16) pixels, that is, when h = + 1/16. This use error matrix MF is applied to the basic area printed in the forward pass. This use error matrix is the same as the use error matrix ME (FIG. 14B) applied to the basic region of the head horizontal tilt deviation described above.

  FIG. 14D shows a use error matrix MG when the head vertical tilt deviation is (+1/16) pixels, that is, when h = + 1/16. This use error matrix MG is applied to the basic area printed in the return pass. The shift amounts of the line matrices LM2 and LM3 of the use error matrices MGa and MGb are the same as the shift amounts of the line matrices LM2 and LM3 of the use error matrices MFa and MFb. The shift directions of the line matrices LM2 and LM3 of the use error matrices MGa and MGb are opposite to the shift directions of the line matrices LM2 and LM3 of the use error matrices MFa and MFb.

  The use error matrix applied to each raster line in the path boundary region is not shown, but the shift target line matrix of the standard error matrix MA so as to cancel the relative shift indicated by the box ZB in FIG. May be created by shifting. Alternatively, the use error matrix for the basic area may be applied without creating a dedicated use error matrix.

  The halftone process in the second embodiment is the same as the halftone process (FIG. 7) in the first embodiment except that the use error matrix is different. For example, in FIG. 7: Step S502, the error matrix acquisition unit M108 acquires the value k representing the head horizontal tilt deviation and the value h representing the head vertical tilt deviation as the positional deviation correction coefficient Pc. In step S504, the error matrix acquisition unit M108 acquires the above-described use error matrix using the correction coefficient Pc.

  According to the second embodiment described above, the halftone processing unit M106 uses the use error matrix adjusted in accordance with the pixel position deviation based on the nozzle deviation such as the head horizontal inclination deviation or the head vertical inclination deviation, and the Perform tone processing. As a result, it is possible to suppress deterioration in print image quality due to pixel position deviation based on nozzle position deviation of the same color nozzle.

  The halftone processing unit M106 performs halftone processing using a use error matrix adjusted according to the nozzle deviation in the main scanning direction represented by the head horizontal inclination deviation. As a result, it is possible to suppress deterioration in print image quality due to pixel position deviation based on nozzle deviation in the main scanning direction. The use error matrix ME (FIG. 14B) used in this case is applied to a raster line printed by forward printing and also applied to a raster line printed by backward printing.

Further, the halftone processing unit M106 performs halftone processing using a use error matrix adjusted in accordance with the nozzle deviation in the facing direction typified by the head vertical tilt deviation. As a result, it is possible to suppress deterioration in print image quality due to pixel position deviation based on nozzle deviation in the facing direction. In this case, a use error matrix MF (FIG. 14C) applied to a raster line printed by forward printing and a use error matrix MG (FIG. 14 (FIG. 14) applied to a raster line printed by backward printing. D)) and used.

C. Third embodiment:
In the third embodiment, halftone processing according to a composite pixel position shift that occurs when both the bidirectional shift in the first embodiment and the nozzle shift in the second embodiment occur will be described. The configuration of the multi-function device 600 is the same as that of the multi-function device 600 (FIG. 1) of the first embodiment. The composite pixel positional shift considered in the present embodiment is a composite shift of the bidirectional shift and the head horizontal tilt shift, and a composite shift of the bidirectional shift and the head vertical tilt shift (also referred to as a second composite pixel positional shift). is there.

C-1: Halftone processing:
In the halftone process in the present embodiment, the error matrix to be used may differ depending on the printing method, as in the first embodiment. In this embodiment, a case where the printing method is the same bidirectional interlaced printing as in the first embodiment (FIG. 5) will be described as an example.

C-1-1: When there is a composite deviation between the bidirectional deviation and the head horizontal inclination deviation:
FIG. 16 is a diagram for explaining a composite shift (also referred to as a first composite pixel position shift) between the bidirectional shift and the head horizontal tilt shift during bidirectional interlaced printing. The left side of FIG. 16 shows the position of the print head 250 in the sub-scanning direction in each pass when bitmap data (not shown) is printed, as in FIG.

  In the present embodiment, the bidirectional displacement of the displacement amount s described in the first embodiment (FIG. 5) and the head horizontal inclination displacement of the displacement amount k described in the second embodiment (FIG. 13) are combined. And The shift amount and the shift direction of the first composite pixel position shift are determined by the sum of the contribution of the bidirectional shift and the contribution of the head horizontal tilt shift. FIG. 16 conceptually shows the contribution of bi-directional deviation in the positional relationship between the dots D1 printed in the forward pass and the dots D2 printed in the return pass. Similarly, the contribution of the head horizontal tilt deviation is also illustrated.

  As can be seen from these conceptual diagrams, the even-numbered raster line (even-numbered line) of the image data is returned to the odd-numbered raster line (odd-numbered line) due to the contribution of bidirectional displacement. It is shifted by s in the (−Dy direction). Then, due to the contribution of the head horizontal deviation, within the range printed in the same pass, the odd-numbered line adjacent to the even-numbered line in the sub-scanning direction is in the forward direction (+ Dy direction). , Shift by k.

  When the sum of these contributions is taken, the relative pixel position shift of the raster line is as shown by a box ZB in FIG. That is, the even-numbered line adjacent to the odd-numbered line in the sub-scanning direction is shifted by s in the backward direction. Then, with respect to the odd-numbered line, the odd-numbered line located in the sub-scanning direction by two lines from the odd-numbered line is shifted by k in the forward direction. Then, the odd-numbered line adjacent to the even-numbered line in the sub-scanning direction is shifted by (s + k) in the forward direction. The even-numbered line located in the sub-scanning direction by two lines from the even-numbered line is shifted by k in the forward direction.

The use error matrix is created by shifting the shift target line matrix of the standard error matrix MA so as to cancel the relative shift shown in the box ZB. FIG. 17 is a diagram illustrating an example of an error matrix used when a composite pixel position shift occurs. FIG. 17B shows a use error matrix MH when the bidirectional deviation is +1 pixel and the head horizontal tilt deviation is +1/16 pixel, that is, when s = + 1 and k = + 1/16. ing. The use error matrix MHa is the second of the standard error matrix MAa.
The line matrix LM2 is obtained by shifting only one pixel in the forward direction (+ Dy direction) and the third line matrix LM3 is shifted by (1/16) pixels in the backward direction (−Dy direction). The use error matrix MHb shifts the second line matrix LM2 of the standard error matrix MAb by (17/16) pixels in the backward direction (−Dy direction), and shifts the third line matrix LM3 in the backward direction (−Dy direction) ( It is obtained by shifting by 1/16) pixel. The usage error matrix MHa is a usage error matrix in the case where pixels on odd-numbered lines are processing target pixels. The usage error matrix MHb is a usage error matrix in the case where pixels on even-numbered lines are processing target pixels.

C-1-2: When a bi-directional deviation and a head vertical inclination deviation occur:
FIG. 18 is a diagram for explaining a composite shift (also referred to as a second composite pixel position shift) of the bidirectional shift and the head vertical tilt shift during bidirectional interlaced printing. The left side of FIG. 17 shows the position of the print head 250 in the sub-scanning direction in each pass when bitmap data (not shown) is printed, as in FIG.

  In this embodiment, the bidirectional displacement of the displacement amount s described in the first embodiment (FIG. 5) and the head vertical inclination displacement of the displacement amount h described in the second embodiment (FIG. 15) are combined. And The shift amount and the shift direction of the second composite pixel position shift are determined by the sum of the contribution of the bidirectional shift and the contribution of the head vertical tilt shift. FIG. 18 conceptually shows the contribution of bi-directional deviation in the positional relationship between the dots D1 printed in the forward pass and the dots D2 printed in the return pass. Similarly, the contribution of the head vertical tilt deviation is also illustrated.

  As can be seen from these conceptual diagrams, the even-numbered raster line (even-numbered line) of the image data is returned to the odd-numbered raster line (odd-numbered line) due to the contribution of bidirectional displacement. It is shifted by s in the (−Dy direction). Then, due to the contribution of the head vertical tilt deviation, within the range printed in the same pass, the second raster line does not deviate from the first raster line, but for the odd line lines except the first line. Thus, the even-numbered line adjacent to the odd-numbered line in the sub-scanning direction is shifted by 2 × (n−1) × h in the backward direction (−Dy direction). Here, n is the nozzle number (nozzle number counted from the upstream side in the sub-scanning direction) of the nozzle 250n that prints the odd line. Further, within the range printed by the same pass due to the contribution of the head vertical tilt deviation, the odd-numbered line adjacent to the even-numbered line in the sub-scanning direction is in the forward direction (+ Dy direction). , (2 × n−1) × h.

  When the sum of these contributions is taken, the relative pixel position shift of the raster line is as shown by a box ZB in FIG. In other words, the even-numbered line adjacent to the odd-numbered line in the sub-scanning direction is shifted by {s + 2 × (n−1) × h} in the backward direction. Then, with respect to the odd-numbered line, the odd-numbered line located in the sub-scanning direction by two lines from the odd-numbered line is shifted by h in the forward direction. Then, the odd-numbered line adjacent to the even-numbered line in the sub-scanning direction is shifted by {s + (2 × n−1) × h} in the forward direction. Then, the even-numbered line located in the sub-scanning direction by two lines from the even-numbered line is shifted by h in the backward direction with respect to the even-numbered line.

The use error matrix is created by shifting the shift target line matrix of the standard error matrix MA so as to cancel the relative shift shown in the box ZB. FIG. 17C shows a use error matrix MH when the bidirectional deviation is +1 pixel and the head vertical inclination deviation is (+1/16) pixel, that is, when s = + 1 and h = + 1/16. Is shown. The use error matrix MJa is used when a raster line pixel printed by the nozzle 1 in the forward pass (for example, pass P1 (F) in FIG. 18) is to be processed. The use error matrix MJb is used when a pixel of a raster line printed by the nozzle 1 is a processing target in a return pass (for example, pass P2 (R) in FIG. 18). The use error matrix MJc is used when a raster line pixel printed by the nozzle 2 in the forward pass is a processing target. The use error matrix MJd is used when a raster line pixel printed by the nozzle 2 in the return pass is a processing target. As described above, in the example of FIG. 18, the use error matrix is different for each raster line included in an area printed by a combination of one forward pass and one return pass.

  According to the third embodiment described above, even if a composite displacement such as the first composite pixel displacement or the second composite pixel displacement occurs, the halftone processing unit M106 does not Halftone processing can be performed using the use error matrix (FIG. 17) adjusted according to the deviation. As a result, it is possible to suppress a decrease in print image quality due to such a composite shift.

D. Fourth embodiment:
D-1: Configuration of sub-scanning section 260 (paper transport section) of the fourth embodiment:
FIG. 19 is an enlarged perspective view of the main scanning unit 240 and the sub-scanning unit 260 (see FIG. 2) in the fourth embodiment. The multi-function device 600 in the fourth embodiment is different from the first embodiment in the configuration of the sub-scanning unit 260 (paper transport mechanism). Specifically, the sub-scanning unit 260 of the MFP 600 according to the fourth embodiment performs conveyance (sub-scanning) of the sheet in a state where the print medium 300 (sheet) is deformed in a wave shape along the main scanning direction. (FIG. 19).

  FIG. 19A also shows a cross section of the platen 265 (cross section perpendicular to the sub-scanning direction + Dx: a cross section at a position between two rollers 261 and 262 (see FIG. 2)). As illustrated, the platen 265 includes a plurality of low support portions 265a and a plurality of high support portions 265b. The high support portion 265b and the low support portion 265a each have a paper support surface along the sub-scanning direction. The paper support surface of the high support portion 265b supports the print medium 300 at a higher position than the paper support surface of the low support portion 265a. The low support portions 265a and the high support portions 265b are alternately arranged along the main scanning direction (+ Dy, −Dy direction).

  A pressing portion 269 is disposed above the paper support surface of the low support portion 265a. In the pressing unit 269, the printing medium 300 sent out by the first roller 261 is bent by the pressing unit 269 and is deformed in a wave shape at a position facing the print head 250 (a position where printing is performed). That is, the print medium 300 is held in a wavy shape along the main scanning direction, which has a valley portion located above the low support portion 265a and a peak portion located above the high support portion 265b. The print medium 300 is conveyed by the sub-scanning unit 260 while being held in such a wavy shape.

  The reason why the print medium 300 is deformed in a wave shape is to prevent the print medium 300 from floating from the platen 265 toward the print head 250 due to the round of the print medium 300.

  FIG. 19B shows the positions of the dots D1 and D2 formed by the inks I1 and I2 ejected from the print head 250. The dots D1 are formed by the first ink I1 ejected from the print head 250 moving in the + Dy direction, and the dots D2 are formed by the second ink I2 ejected from the print head 250 moving in the -Dy direction.

  The gaps GP3 and GP4 in FIG. 19B indicate gaps between the nozzle surface np and the print medium 300. The gap GP3 is a gap at the peak portion Pyb of the print medium 300 that is deformed in a wave shape. The gap GP4 is a gap in the valley Pya of the print medium 300 that is deformed in a wave shape.

  As shown in the figure, the area on the print medium 300 is divided into a first type area A1 and a second type area A2 that are alternately arranged along the main scanning direction. 1st type area | region A1 is an area | region including the trough part Pya pressed by the press part 269. FIG. 2nd type area | region A2 is an area | region containing the peak part Pyb supported by the high support part 265b. The first type region A1 is, for example, a region where the gap between the nozzle surface np and the print medium 300 is larger than the intermediate value between the gap GP3 and the gap GP4. In addition, the second type region A2 is a region where the gap between the nozzle surface np and the print medium 300 is equal to or less than the intermediate value between the gap GP3 and the gap GP4, for example.

  The ejection timings of the inks I1 and I2 are adjusted on the assumption that printing is performed on the horizontal plane SH passing through the peak portion Pyb. Accordingly, the pixel position deviation is relatively small in the second type region A2 in the vicinity of the peak portion Pyb, but the pixel position deviation is relatively large in the first type region A1 in the vicinity of the valley portion Pya. FIG. 19B shows dots D1 and D2 formed by the inks I2 and I2 ejected to form dots in the valley Pya. The gap GP4 of the valley portion Pya is larger than the gap GP3 of the peak portion Pyb. As a result, when printing is performed on the first type area A1 including the valley Pya, the landing positions (dot formation positions) of the inks I1 and I2 are shifted from the target position Pya (FIG. 19B).

  As can be seen from the above description, in the fourth embodiment, even in the same raster line, the bidirectional displacement (see the first embodiment) differs depending on the position in the main scanning direction. That is, the bidirectional displacement in the first type region A1 is larger than the bidirectional displacement in the second type region A2.

  FIG. 20 is a diagram for explaining halftone processing in the fourth embodiment. FIG. 20A shows a target image ID represented by target image data (image data to be printed). The target image data represents an image having a uniform gradation value. FIG. 20B shows a print image IP obtained by printing the target image data without performing the halftone process in the present embodiment. As shown in the drawing, graininess unevenness occurs between the first type region A1 and the second type region A2. As described above, the difference in graininess between the first type region A1 and the second type region A2 is caused by the difference in bidirectional displacement between the first type region A1 and the second type region A2. Because it occurs.

D-2: Halftone processing:
FIG. 20C shows the contents of the correction coefficient data 138 in the fourth embodiment. In the correction coefficient data 138 in this embodiment, a plurality of types of reference correction coefficients Pref associated with positions in the main scanning direction on the print medium 300 are described. The positions of the plurality of valley portions Pya described with reference to FIG. 19B are Pya1, Pya2,..., Pya6 from the upstream side in the + Dy direction to the downstream side of the print medium 300 (FIG. 20). (B)). Further, the positions of the plurality of peak portions Pyb are Pyb1, Pyb2,..., Pyb7 from the upstream side to the downstream side in the + Dy direction of the print medium 300 (FIG. 20B). The positions of the valleys Pya and the peaks Pyb are recorded, for example, in units of pixels of the print image. Then, the reference correction coefficient Pref (1.0 in this embodiment) corresponding to the bidirectional displacement in the valley Pya is associated with the positions Pya1, Pya2,..., Pya6 of the valley Pya. On the other hand, the reference correction coefficient Pref (0.0 in the present embodiment) corresponding to the bi-directional deviation in the peak portion Pyb is associated with the above-described positions Pyb1, Pyb2, ..., Pyb7 of the peak portion Pyb. ing. That is, in this embodiment, it is assumed that there is no bidirectional shift at the peak portion Pyb position in the print medium 300, and that the maximum bidirectional shift occurs at the valley portion Pya position in the print medium 300. Correction coefficient data 138 is described.

The outline of the halftone process in the fourth embodiment is the same as the halftone process in the first embodiment shown in FIG. However, in step S502 of FIG. 7, the error matrix acquisition unit M108 acquires the correction coefficient Pc corresponding to the position of the processing target pixel in the main scanning direction, using the correction coefficient data 138 shown in FIG. Specifically, the error matrix acquisition unit M108 acquires the correction coefficient Pc by linear interpolation using the reference correction coefficient Pref described in the correction coefficient data 138.

  FIG. 20D conceptually shows the correction coefficient Pc acquired by the error matrix acquisition unit M108. Thus, for example, at an intermediate position between the valley Pya and the peak Pyb, the reference correction coefficient Pref (1.0) associated with the valley Pya and the reference correction coefficient Pref associated with the peak Pyb. A value intermediate between (0.0) is acquired as the correction coefficient Pc. As described with reference to FIG. 19B, the position in the main scanning direction is such that the gap between the nozzle surface np and the print medium 300 is maximum at the valley Pya and minimum at the peak Pyb. Depending on the, it is continuously changing. Therefore, it can be said that the correction coefficient Pc acquired in the present embodiment corresponds to the size of the gap between the nozzle surface np and the print medium 300.

  As a result of acquiring such a correction coefficient Pc, in this embodiment, the error matrix acquisition unit M108 shifts the shift target line matrix in step 504 of FIG. 7 according to the position of the processing target pixel in the main scanning direction. However, different usage error matrices can be obtained.

  According to the fourth embodiment described above, as a use error matrix used for bidirectional printing, a plurality of pixels adjusted according to different pixel position deviations (bidirectional deviations) according to positions in the main scanning direction on the printing medium 300 are used. Error matrix set (a set of use error matrices having different correction coefficients Pc) is used. Then, the halftone processing unit M106 executes halftone processing while changing the error matrix set to be used among these use error matrix sets according to the position of the processing target pixel in the main scanning direction.

  As a result, even if the pixel position shift (for example, bidirectional shift) differs depending on the position of the processing target pixel in the print medium 300 in the main scanning direction, the print image quality is reduced due to the pixel position shift. It can suppress that it falls.

  More specifically, the MFP 600 according to the present exemplary embodiment performs sub-scanning in a state where the print medium is deformed so that the gap between the print medium 300 and the nozzles 250n changes along the main scanning direction. A sub-scanning unit 260 is provided. In a plurality of error matrix sets (sets of use error matrices having different correction coefficients Pc), the shift amount of the shift target line matrix differs for each error matrix set according to the size of the gap.

  As a result, even when the printing execution unit 200 performs sub-scanning in a state where the gap between the printing medium 300 and the nozzles 250n is deformed so as to change along the main scanning direction, It is possible to suppress a decrease in print image quality due to a pixel position shift based on the deformation. For example, the granularity unevenness between the first type area A1 and the second type area A2 shown in FIG. 20B can be reduced, and the image quality of the printed image can be improved.

E. Example 5:
In the first to fourth embodiments, the halftone processing unit M106 performs halftone processing using the error diffusion method described with reference to FIGS. 6 and 7, but instead of this, error collection is performed. You may adopt the law.

FIG. 21 is a diagram showing an outline of the error collection method. In the error diffusion method, in step S514 of FIG. 7, the halftone processing unit M106 distributes the peripheral pixels (unprocessed pixels that are not processed) of the calculated target error value Ea and the distribution ratio for each distribution destination. Is determined according to the use error matrix. Then, the halftone processing unit M106 stores each peripheral pixel in the error buffer EB. In the error collection method, the halftone processing unit M106 stores the calculated target error value Ea in the error buffer EB as an error value generated in the processing target pixel. Then, when the distribution error value Et to be added to the correction input value Vin2 is determined when executing step S508 of FIG. 7 for the pixel whose processing order is later, the halftone processing unit M106 performs error collection. A distribution error value Et is calculated using an error matrix (described later). Specifically, the halftone processing unit M106 uses the error collection error matrix, and target error values of peripheral pixels that are already processed pixels (determined pixels for which dot data has been determined) that are already processed. From Ea, errors for the amount distributed to the processing target pixel are collected to calculate a distribution error value Et. Although there are differences in the manner in which the target error value Ea is stored in the error buffer EB in step S514 and the manner in which the distribution error value Et is obtained in step S508, the error diffusion method and the error collection method are basically the same. Is the same halftone processing method.

  22 and 23 are diagrams showing examples of use error matrices for the error collection method. FIG. 22A shows standard error matrices MKa and MKb for error collection. This standard error matrix MK is substantially equivalent to the standard error matrix MA for error diffusion (FIG. 8A), and is designed to realize a suitable image quality when no pixel position deviation occurs. Error matrix. This standard error matrix MK includes line matrices LM1 to LM3 for three rows. The first line matrix LM1 defines the error distribution ratio for the peripheral pixels located on the target raster line where the processing target pixel is located. The second line matrix LM2 in the error collection error matrix defines the error distribution ratio for the peripheral pixels located in the first reverse proximity raster line of the target raster line. The third line matrix LM3 in the error collection error matrix defines the error distribution ratio for the peripheral pixels located in the second reverse proximity raster line of the target raster line. Here, the first reverse proximity raster line is a raster line adjacent to the target raster line in the reverse direction to the first proximity raster line described above. That is, the first reverse proximity raster line is a raster line adjacent to the target raster line in the direction opposite to the sub-scanning direction (−Dx direction). The second reverse proximity raster line is a raster line that is located in the opposite direction (−Dx direction) of the sub-scanning direction by two rows from the target raster line.

  Similarly to the error diffusion use error matrix, the error collection use error matrix is created by shifting the shift target line matrix of the standard error matrix MK according to the relative deviation. The shift target line matrix is a line matrix in which a relative shift occurs in the corresponding raster line (first reverse proximity raster line, second reverse proximity raster line) of the second line matrix LM2 and the third line matrix LM3. . The relative shift is a pixel position shift with respect to the processing target pixel of the target raster line. The shift of the shift target line matrix is executed so as to cancel the relative shift, similarly to the creation of the error diffusion use error matrix. That is, the shift target matrix is shifted with respect to the first line matrix LM1 by the same amount as the relative shift amount in the direction opposite to the relative shift.

  The error collection use error matrices MLa and MLb shown in FIG. 22B are suitable error matrices when a bidirectional shift of s = + 1 occurs, and the error diffusion use error matrices MBa and MBb. (FIG. 8B) is almost equivalent. The error collection use error matrices MMa and MMb shown in FIG. 22C are suitable error matrices when a bidirectional shift of s = + 0.5 occurs, and the error diffusion use error matrix MC2a. MC2b (FIG. 8C) is almost equivalent. The error collection use error matrices MNa and MNb shown in FIG. 23 are suitable error matrices when the head horizontal tilt deviation of k = + 1/16 occurs, and the error diffusion use error matrices MEa and MEb. (FIG. 14B) is almost equivalent.

  In the fifth embodiment described above, the halftone processing unit M106 executes the same halftone processing as in the first to fourth embodiments described above using the error collection method instead of the error diffusion method. According to the fifth embodiment, the same operations and effects as the first to fourth embodiments described above are achieved. That is, even when various pixel position shifts occur, it is possible to suppress a decrease in print image quality due to the pixel position shifts.

F. Example 6:
In the sixth embodiment, a configuration and processing for generating correction coefficient data 138 will be described. FIG. 24 is a flowchart of the evaluation image printing process. The evaluation image printing process is executed, for example, when the control device 100 of the multifunction peripheral 600 receives a print instruction from the user. This print instruction may be, for example, a print instruction that directly instructs printing of the evaluation image, or may be an image quality improvement instruction that requests improvement of image quality such as graininess.

  When the evaluation image printing process is started, in first step S1010, the evaluation image data generation unit M110 of the print data generation unit M100 acquires CMYK evaluation image data. The CMYK evaluation image data is bitmap data composed of CMYK pixel data, and is image data representing the evaluation print image IM1 including a plurality of evaluation images. FIG. 25 is a schematic diagram showing the contents of the evaluation print image IM1.

  The evaluation print image IM1 includes, as a plurality of evaluation images, a cyan (C) single color patch group CG, a magenta (M) single color patch group MG, and a yellow (Y) single color patch group YG. And a black (K) single color patch group KG. In the plurality of color patches included in the color patch groups CG, MG, YG, and KG for each color, the corresponding component values of C, M, Y, and K have specific values, and the other component values are zero. It is represented by monochromatic pixel data. The specific value is set to, for example, a value (for example, 128 which is an intermediate value of 256 gradations) representing a density suitable for evaluation of graininess. Each color patch group CG, MG, YG, KG has three patch rows arranged along the main scanning direction (+ Dy, -Dy direction). The three patch rows for each color are the first type of evaluation patches CPb, MPb, YPb, KPb, the second type of evaluation patches CPf, MPf, YPf, KPf, and the third type of evaluation patch CPr, MPr, YPr, and KPr rows. The first type of evaluation patches CPb, MPb, YPb, and KPb are evaluation patches for evaluating bidirectional displacement. The second type of evaluation patches CPf, MPf, YPf, and KPf are evaluation patches for evaluating head tilt deviation during forward printing. The third type of evaluation patches CPr, MPr, YPr, and KPr are evaluation patches for evaluating the head tilt deviation during the return pass printing.

  In step S1020 of FIG. 24, the evaluation image data generation unit M110 acquires a positional deviation correction coefficient Pc used for each evaluation patch. Specifically, the evaluation image data generation unit M110 includes correction coefficients Pcb for the first type of evaluation patches CPb, MPb, YPb, and KPb, and second and third types of evaluation patches CPf, MPf, YPf, The correction coefficient Pcg for KPf, CPr, MPr, YPr, KPr is acquired. The correction coefficient Pcb is a position deviation correction coefficient for evaluating bidirectional deviation, and the correction coefficient Pcg is a correction coefficient for evaluating head tilt deviation.

  In step S1030, the evaluation image data generation unit M110 executes halftone processing while changing the correction coefficient Pc to be used for each evaluation patch based on the acquired correction coefficients Pcg and Pcb. Specifically, the evaluation image data generation unit M110 determines the value of the correction coefficient Pc to be used and the type of the error matrix for each evaluation patch. The halftone processing unit M106 creates an error matrix of the determined type using the determined correction coefficient Pc, and executes halftone processing for each evaluation patch. As a result, CMYK pixel data representing each evaluation patch is converted into dot data.

  FIG. 26 is a diagram for explaining a use error matrix used for generating evaluation image data. FIG. 26A is an example of a use error matrix used in the halftone process for the first type of evaluation patch. A use error matrix used for halftone processing for a certain evaluation patch is also referred to as a use error matrix corresponding to the evaluation patch. In this example, use error matrices MOa and MOb when the correction coefficient Pc = Pcb are shown. In FIG. 26, the shift amounts are described in the line matrices LM2 and LM3 of each use error matrix. For example, it can be seen that the use error matrix MOa is created by shifting the line matrix LM2 of the standard error matrix MAa (FIG. 8A) by Pcb in the forward direction. Further, it can be seen that the use error matrix MOb is created by shifting the line matrix LM2 of the standard error matrix MAb (FIG. 8A) by Pcb in the backward direction. As can be seen from this shift mode, the usage error matrices MOa and MOb are the type of usage error matrix used when a bidirectional shift occurs. The use error matrices MOa and MOb are also referred to as bidirectional shift error matrices corresponding to the correction coefficient Pc = Pcb.

  FIG. 26B shows an example of a use error matrix associated with the second type and third type evaluation patches. In this example, use error matrices MPa and MPb when the correction coefficient Pc = Pcg are shown. The use error matrix MPa shifts the line matrix LM2 of the standard error matrix MAa (FIG. 8A) by Pcg in the forward direction, and shifts the third line matrix LM3 of the standard error matrix MAa by 2 × Pcg in the forward direction. It can be seen that It can be seen that the use error matrix MPb is obtained by similarly shifting the standard error matrix MAb (FIG. 8A). As can be seen from this shift mode, the use error matrices MPa and MPb are use error matrices of the type used when head tilt deviation occurs. The use error matrices MPa and MPb are also referred to as a head tilt deviation error matrix corresponding to the correction coefficient Pc = Pcg.

  The values shown in each evaluation patch (rectangular square) in the evaluation print image IM1 shown in FIG. 25 indicate the value of the correction coefficient Pc used to create the use error matrix corresponding to the evaluation patch. ing. This value is also called a correction coefficient corresponding to the evaluation patch. For example, seven first-type evaluation patches CPb are arranged in a row of the first-type cyan evaluation patch CPb (the uppermost row in FIG. 25). These seven first-type evaluation patches CPb correspond to seven types of error matrices for bidirectional deviation, each having a different value of the corresponding correction coefficient Pc. The misregistration correction coefficients Pc corresponding to the seven types of bidirectional misalignment error matrices are seven values −3Pcb to + Pcb with Pcb as the step size and centering on zero.

  In the row of the second type evaluation patch CPf of cyan (FIG. 25: second row from the top) and the row of the third type evaluation patch CPr of cyan (FIG. 25: third row from the top), Seven evaluation patches CPf and CPr are arranged side by side. The seven evaluation patches CPf and CPr in each row correspond to seven types of head tilt deviation error matrices having different values of the corresponding correction coefficient Pc. The misregistration correction coefficients Pc corresponding to the seven types of head tilt misalignment error matrices are seven types of −3 Pcg to + Pcg centered on zero with Pcg as the step size. Here, the size of Pcb or Pcg as the step size may be smaller than one print pixel (one dot). When the size expressed in pixel units is 0 <Pcb and Pcg <1, the use error matrix corresponding to the correction coefficient Pc is created by using the decimal shift method described in the first embodiment. can do.

  Although the details of the configuration of the cyan color patch group CG have been described, the configurations of the color patch groups MG, YG, and KG of other colors are the same as the configuration of the cyan color patch group CG, as can be seen from FIG. It is.

  When the entire dot data of the evaluation print image IM1 is generated by the halftone process, in step S1040, the evaluation image data generation unit M110 generates evaluation image print data for printing the evaluation print image IM1. The evaluation image data generation unit M110 generates evaluation image print data so that each row of the first type of evaluation patches CPb, MPb, YPb, and KPb is printed by bidirectional interlaced printing (FIG. 5). Further, the evaluation image data generation unit M110 prints each row of the second type of evaluation patches CPf, MPf, YPf, KPf by one-way printing only by the forward printing, and outputs the third type of evaluation patches CPr, MPr, YPr, Evaluation image print data is generated so that each line of KPr is printed by one-way printing using only backward printing.

  Here, the evaluation image data generation unit M110 includes a line printed by bidirectional interlaced printing (for example, a line of the first type evaluation patch CPb of cyan) for one forward path and one backward path. The evaluation image print data is generated so that the entire line is printed in two passes consisting of the combinations. Here, as in the combination of the forward path P1 (F) and the backward path P2 (R) in FIG. 5, the amount of sub-scanning between one forward path and one backward path is sub-scanning. The amount is the same as the direction dot pitch (in this example, half the nozzle pitch). As described above, by using the minimum amount of sub-scanning, the evaluation patch can be widened in the sub-scanning direction even though the evaluation patch is printed with a small number of main scans. That is, it is possible to print an evaluation patch that allows easy evaluation of image quality with a small number of main scans.

  In step S1050, the evaluation image data generation unit M110 supplies the generated evaluation image print data to the print execution unit 200. The print execution unit 200 prints the evaluation image print data, and prints the evaluation print image IM1 on the print medium 300.

  When the evaluation image printing process ends, that is, when the evaluation print image IM1 is printed, the print data generation unit M100 executes the print setting process. FIG. 27 is a flowchart illustrating processing steps of the print setting process.

  In step S2010, the image quality evaluation acquisition unit M112 of the print data generation unit M100 acquires the user's visual evaluation result for the printed evaluation print image IM1. The visual evaluation result is acquired for each row of the evaluation patch in the evaluation print image IM1, that is, for each printing method and each ink color. For example, the image quality evaluation acquisition unit M112 receives, via the operation unit 170, the patch number of the evaluation patch that the user feels has good image quality (for example, granularity) for each row of the evaluation patch.

  In step S2020, the print setting determination unit M114 determines an appropriate value of the misregistration correction coefficient Pc for each printing method and for each ink color based on the acquired visual evaluation result. Specifically, the print setting determination unit M114 determines the correction coefficient Pc (see FIG. 25) corresponding to the evaluation patch designated by the user as having good image quality as the appropriate correction coefficient Pc. For example, as an appropriate correction coefficient Pc for cyan printing, a correction coefficient Pc suitable for bidirectional printing based on the evaluation result for the first type evaluation patch CPb is used as the evaluation result for the second type evaluation patch CPf. Based on this, the correction coefficient Pc suitable for the forward printing is determined, and the correction coefficient Pc suitable for the backward printing is determined based on the evaluation result of the third type evaluation patch CPr. The print setting determining unit M114 stores the determined plural types of correction coefficients Pc in the nonvolatile memory 130 as correction coefficient data 138. The correction coefficient Pc determined here is used, for example, to create a use error matrix when performing the above-described bidirectional interlaced printing or bidirectional noninterlaced printing.

According to the sixth embodiment described above, image data (for example, bitmap data BD)
A plurality of evaluation patches capable of evaluating a change in print image quality (for example, a change in graininess) due to a shift (pixel position shift) in the main scanning direction pixel position in the print medium 300 with respect to the main scanning direction pixel position in Can be easily printed by the print execution unit 200. Specifically, the use error matrix is changed by changing the value of the misregistration correction coefficient Pc without performing complicated control such as adjustment of ink ejection timing, and an appropriate evaluation print image IM1 is obtained. Can be printed.

  The use error matrix is obtained by shifting the shift target line matrix of the standard error matrix MA in the main scanning direction based on the correction coefficient Pc. As a result, the print execution unit 200 can easily cause the print execution unit 200 to print the evaluation print image IM1 that can evaluate the change in the print image quality caused by the positional deviation between the pixel of the target raster line and the pixel of the other raster line. Can do.

  Further, even when the value of Pcb or Pcg, which is the step size of the correction coefficient Pc, is a decimal number (0 <Pcb, Pcg <1) in units of pixels, the use error is determined by using the decimal shift method. A matrix can be created. As a result, it is possible to easily cause the print execution unit 200 to print the evaluation print image IM1 that can evaluate the change in the print image quality caused by the pixel position shift smaller than one pixel.

  Here, the first type of evaluation patch is an evaluation patch for evaluating bidirectional printing. In the sixth embodiment, an error matrix set such as a set of use error matrix MOa and use error matrix MOb is used for the halftone process for the first type of evaluation patch. The use error matrix MOa is an example of a forward error matrix used when the raster line where the processing target pixel is located is printed by forward printing, and the usage error matrix MOb is a return path where the raster line where the processing target pixel is located. It is an example of a return path error matrix used when printing is performed. The shift direction of the shift target line matrix (second line matrix LM2) of the use error matrix MOa is opposite to the shift direction of the shift target line matrix of the use error matrix MOa (FIG. 26). With this configuration, it is possible to easily cause the print execution unit 200 to print the first type of evaluation patch that can evaluate the change in print image quality caused by the pixel position shift in bidirectional printing.

  Furthermore, the second type evaluation patch is printed using unidirectional printing using only forward printing, and the third type evaluation patch is printed using unidirectional printing using only backward printing. For these evaluation patches, halftone processing using a head tilt deviation error matrix MP (FIG. 26B) is used. As a result, it is possible to easily cause the print execution unit 200 to print the evaluation print image IM1 that can evaluate the change in the print image quality caused by the pixel position shift based on the nozzle position shift of the same color nozzle such as the head tilt shift. it can.

  Also, the degree of pixel position deviation due to head tilt deviation may differ between forward pass printing and return pass printing. This can occur when a head tilt shift in which the head horizontal tilt shift (FIG. 13) and the head vertical tilt shift (FIG. 15) are combined occurs. In this embodiment, the evaluation includes both the second type patch printed by unidirectional printing using only forward printing and the third type patch printed by unidirectional printing using only backward printing. The print image IM1 is printed. As a result, it is possible to appropriately evaluate a change in print image quality caused by a pixel position shift based on a nozzle position shift in the main scanning direction such as a head tilt shift.

Furthermore, in this embodiment, an image quality evaluation for the print result of the evaluation print image IM1 is acquired, and correction coefficient data 138 (correction coefficient Pc) used for subsequent printing is created using the acquired image quality evaluation. ing. Therefore, it is possible to appropriately determine the correction coefficient data 138 using the evaluation print image IM1. As a result, in subsequent printing, it is possible to suppress a decrease in print image quality due to pixel position shift.

  Furthermore, in the evaluation print image IM1, evaluation patches to be printed by the same printing method are arranged side by side along the main scanning direction. That is, evaluation patches printed by different printing methods are not mixed in the same row along the main scanning direction. As a result, a plurality of evaluation patches can be efficiently printed by the same main scanning. As a result, the evaluation print image IM1 can be easily printed in a short time. In this embodiment, the change in image quality characteristics between the same type of evaluation patches is realized by changing the correction coefficient Pc corresponding to the use error diffusion matrix. This fact also contributes to printing the evaluation print image IM1 in a short time. For example, if the image quality characteristics of the evaluation patch are changed by changing the ejection timing of ink during main scanning to cause pixel position deviation, the same is true even if the evaluation patches are arranged in the main scanning direction. Printing may not be possible with main scanning. This is because it is difficult to accurately change the ink ejection timing during the same main scan, and thus there is a possibility that a plurality of evaluation patches must be printed in separate main scans.

G. Seventh embodiment:
An evaluation print image IM2 having a different aspect from the evaluation print image IM1 (FIG. 25) will be described as a seventh example. FIG. 28 is a schematic diagram showing the contents of the evaluation print image IM2. This evaluation print image IM2 is used in a printing apparatus including the sub-scanning unit 260 (FIG. 19) described in the fourth embodiment. That is, it is used in a configuration that includes a sub-scanning unit 260 that performs sheet conveyance (sub-scanning) in a state where the print medium 300 is deformed in a wave shape along the main scanning direction.

  In the evaluation print image IM2, the first patch group PG1 is disposed in each of the plurality of first type regions A1 described with reference to FIG. 20, and the second patch group PG2 is provided in each of the plurality of second type regions A2. Has been placed. The first patch group PG1 includes seven first type evaluation patches KPb arranged in the sub-scanning direction. The second patch group PG2 includes seven first type evaluation patches KPb arranged in the sub-scanning direction.

  The seven first type evaluation patches KPb of the first patch group PG1 correspond to seven types of error matrices for bidirectional deviation, each having a different value of the corresponding correction coefficient Pc. The misregistration correction coefficients Pc corresponding to these seven types of bidirectional misalignment error matrices are seven types of values (1-3Pcb) to (1 + 3Pcb) with Pcb as the step size and centering on 1.0. Similarly, the seven first type evaluation patches KPb of the second patch group PG2 correspond to seven types of error matrices for bidirectional deviation, each having a different value of the corresponding correction coefficient Pc. The misregistration correction coefficients Pc corresponding to these seven types of bidirectional misalignment error matrices are seven values −3Pcb to + 3Pcb with Pcb as the step size and centering on zero. That is, the range of the correction coefficient Pc corresponding to the first type of evaluation patch KPb in the first patch group PG1 is different from the range of the correction coefficient Pc corresponding to the first type of evaluation patch KPb in the second patch group PG2. Yes. The correction coefficient Pc corresponds to the shift amount of the second line matrix LM2 in the bidirectional shift error matrix.

  As described with reference to FIG. 19B, the second type region A2 has a relatively small gap between the nozzle surface np and the print medium 300 (a range close to the gap GP3 shown in FIG. 19B). ) Area. The second type region A2 is a region having a relatively large gap (a range close to the gap GP4 shown in FIG. 19B).

As described with reference to FIG. 19B, the ink ejection timing during the main scan is adjusted based on the second type region A2 including the peak portion pyb, and thus occurs in the second type region A2. It can be easily estimated that the bidirectional displacement (pixel positional displacement) is close to zero and the bidirectional displacement occurring in the first type region A1 is relatively large. For this purpose, the first-type evaluation patch KPb of the second patch group PG2 arranged in the second-type region A2 is associated with the error matrix for bidirectional deviation having a shift amount in the range centered on zero. Yes. Then, the first type evaluation patch KPb of the first patch group PG1 arranged in the first type region A1 corresponds to a value having a relatively large shift amount (in this embodiment, the correction coefficient Pc = 1.0). An error matrix for bidirectional deviation in a range centered on (shift amount) is associated.

  By performing the image quality evaluation using the evaluation print image IM2, it is possible to obtain an appropriate correction coefficient Pc in each of the plurality of first type regions A1 and the plurality of second type regions A2. Using the obtained correction coefficient Pc, appropriate correction coefficient data 138 (see FIG. 20C) can be created.

  By using this evaluation print image IM2, when the print execution unit 200 performs sub-scanning in a state where the gap between the print medium 300 and the nozzle is changed so as to change along the main scanning direction, It is possible to evaluate a change in print image quality due to a bidirectional shift (pixel position shift) based on deformation of the print medium 300.

  Further, the shift amount of the corresponding error matrix is between the first type evaluation patch KPb arranged in the first type region A1 and the first type evaluation patch KPb arranged in the second type region A2. The range of the shift amount of the corresponding error matrix is different from each other. As a result, an increase in the number of first type evaluation patches KPb can be suppressed.

H. Variations:
(1) In the seventh embodiment, in the print setting process described with reference to FIG. 27, the image quality evaluation acquisition unit M112 and the print setting determination unit M114 replace the steps S2010 and S2020 described above with a broken line in FIG. Steps S2040 to S2060 may be executed.

  In step S2040, the image quality evaluation acquisition unit M112 acquires scan data generated by reading a print document on which the evaluation print image IM1 is printed using the scanner unit 400 (FIG. 1).

粒状性評価値ＧＥが大きいほど、印刷画像が粗い。ここで、ｕは空間周波数（単位は、サイクル／ミリメートル）であり、ＷＳ（ｕ）は、印刷画像の空間周波数スペクトル（ウイナースペクトラム）である。ＶＴＦ（ｕ）は、視覚の空間周波数特性である。Ｄは、印刷画像の平均濃度である。画質評価取得部Ｍ１１２は、スキャンデータをグレースケールに変換した後、スキャンデータに対して二次元のＦＦＴ（Fast Fourier Transform）を実行する。画質評価取得部Ｍ１１２は、次に、二次元ＦＦＴの結果を、極座標系で表されたデータに変換する。画質評価取得部Ｍ１１２は、極座標系で表されたデータを、角度についての平均化を利用して、一次元化することによって、空間周波数スペクトルＷＳ（ｕ）
を得る。なお、空間周波数スペクトルＷＳ（ｕ）と粒状性評価値ＧＥとの算出方法は、例えば、以下の文献「日本写真学会誌 第６０巻 第６号（１９９７年） インクジェットプリンタの画質評価 藤野真」で説明されている。 In step S2050, the image quality evaluation acquisition unit M112 analyzes the acquired scan data and acquires the granularity evaluation value GE. The graininess evaluation value GE is a numerical value of the graininess (roughness of the image) of the printed image, and is calculated according to the following formula.

The larger the graininess evaluation value GE, the coarser the printed image. Here, u is a spatial frequency (unit: cycle / millimeter), and WS (u) is a spatial frequency spectrum (Wiener spectrum) of a printed image. VTF (u) is a visual spatial frequency characteristic. D is the average density of the printed image. The image quality evaluation acquisition unit M112 performs two-dimensional FFT (Fast Fourier Transform) on the scan data after converting the scan data to gray scale. Next, the image quality evaluation acquisition unit M112 converts the result of the two-dimensional FFT into data expressed in a polar coordinate system. The image quality evaluation acquisition unit M112 makes the spatial frequency spectrum WS (u) by converting the data expressed in the polar coordinate system into a one-dimensional data by using the averaging of the angles.
Get. Note that the calculation method of the spatial frequency spectrum WS (u) and the graininess evaluation value GE is, for example, in the following document “Journal of the Photographic Society of Japan Vol. 60, No. 6 (1997) Image Quality Evaluation of Inkjet Printer Makoto Fujino” Explained.

  In step S2060, the print setting determination unit M114 determines an appropriate value of the misregistration correction coefficient Pc based on the acquired granularity evaluation value GE. Specifically, the print setting determination unit M114 specifies an evaluation patch having the smallest granularity evaluation value GE among evaluation patch groups of the same color and the same type. Then, the correction coefficient Pc (see FIG. 25) corresponding to the identified evaluation patch is determined as an appropriate correction coefficient Pc.

  According to this modification, the scan data is analyzed to determine the image quality evaluation, so that the burden on the user can be reduced.

(2) The above embodiments can be appropriately combined. For example, the print execution unit 200 performs the first printing mode for executing bidirectional interlaced printing (two-pass printing) in the first embodiment and the bidirectional non-interlaced printing (one-pass printing) in the second embodiment. The second print mode to be executed may be executable. In this case, the print data generation unit M100, for example, as a use error matrix, an error matrix for bidirectional displacement (FIG. 8) used in the first print mode and a head tilt displacement used in the second print mode. Error matrix (FIG. 14). Bidirectional misalignment occurs between odd-numbered lines and even-numbered lines in the first print mode, but not in the second print mode. In such a case, the specific line matrix of the use error matrix for the first print mode and the second print mode are set by the difference in pixel position difference between the first print mode and the second print mode. It is preferable that the corresponding use error matrix is shifted from the corresponding line matrix in the main scanning direction.

  According to the above configuration, in each of the first print mode and the second print mode in which the pixel position shifts are different from each other, it is possible to suppress a decrease in print image quality due to the pixel position shift.

(3) The print execution unit 200 in the above embodiment is an ink jet printing mechanism, but may be a laser printing mechanism instead. In particular, when a multi-beam laser capable of simultaneously irradiating a photoconductor with a laser beam on a plurality of main scanning lines is used, the position of a pixel on the other main scanning line with respect to one pixel of the plurality of main scanning lines. May deviate in the main scanning direction. In such a case, as in the above embodiment, halftone processing is performed using an error matrix adjusted according to the pixel position deviation, thereby suppressing deterioration in the print image quality due to the pixel position deviation. can do.

(4) In the sixth embodiment, the print setting determination unit M114 determines the value of the correction coefficient Pc described in the correction coefficient data 138 as the print setting. However, the present invention is not limited to this. Since the correction coefficient Pc corresponds to the direction and magnitude of the pixel position deviation, the print setting determination unit M114 specifies the evaluation patch with the best image quality (for example, the smallest granularity evaluation value GE). The size and direction of the pixel position deviation occurring in the print execution unit 200 can be acquired. The print setting determination unit M114 may determine, as print settings, a change in ink ejection timing according to the acquired pixel position deviation.

(5) In each of the above embodiments, the error matrix acquisition unit M108 acquires the use error matrix by shifting the line matrix of the standard error matrix MA using the correction coefficient Pc. Instead, the line matrix is shifted in advance, that is, a use error matrix created in advance may be stored in the nonvolatile memory 130 as the error matrix data 136.

(6) In the first embodiment, two-pass bi-directional interlaced printing has been described as an example of a mode in which bi-directional misalignment occurs. However, the same method can be used for any other printing method in which bi-directional misalignment occurs. The image quality can be improved. For example, in consideration of the number of passes such as 3-pass printing, 4-pass printing, and 8-pass printing, and sub-scanning modes such as uniform feed and irregular feed, the first adjacent raster line and the second adjacent raster for the pixel to be processed A use error matrix corresponding to the printing method may be created by identifying the pixel pixel position shift and shifting the corresponding line matrix in accordance with the position shift.

(7) In the above embodiments, the control device 100 and the print execution unit 200 may be realized as independent devices housed in different housings. In each of the above embodiments, a plurality of computers that can communicate with each other via a network share a part of the functions required for image processing for printing, and as a whole, the functions of image processing for printing are provided. (A technology using such a computer system is also called cloud computing). For example, a process of analyzing the scan data of the evaluation print image IM1 and calculating the granularity evaluation value GE (the modified example (1)) or the like is a server (for example, a so-called so-called communicator) that is communicably connected to the MFP 600. (Cloud server).

(8) In each of the above embodiments, a part of the configuration realized by hardware may be replaced with software, and conversely, part or all of the configuration realized by software is replaced with hardware. You may do it.

  In addition, when part or all of the functions of the present invention are realized by software, the software (computer program) can be provided in a form stored in a computer-readable recording medium. “Computer-readable recording media” are not limited to portable recording media such as memory cards and CD-ROMs, but are connected to internal storage devices in computers such as various RAMs and ROMs, and computers such as hard disk drives. It also includes an external storage device.

  As mentioned above, although this invention was demonstrated based on the Example and the modification, Embodiment mentioned above is for making an understanding of this invention easy, and does not limit this invention. The present invention can be changed and improved without departing from the spirit and scope of the claims, and equivalents thereof are included in the present invention.

DESCRIPTION OF SYMBOLS 100 ... Control apparatus, 110 ... CPU, 120 ... Volatile memory, 130 ... Non-volatile memory, 132 ... Program, 136 ... Error matrix data, 138 ... Correction coefficient data , 170 ... operation section, 180 ... display section, 190 ... communication section, 200 ... print execution section, 210 ... control circuit, 216 ... sub-scan control section, 240 ... Main scanning unit, 242 ... Moving motor, 244 ... Support shaft, 250 ... Print head, 250n ... Nozzle, 260 ... Sub-scanning unit, 300 ... Print medium, 400 ... Scanner unit, 600 ... MFP, M100 ... Print data generation unit, M102 ... Image data acquisition unit, M104 ... Multi-tone value correction unit, M106 ... Halftone processing unit, M108. .. Error matrix acquisition unit, M110 ... evaluation image data generation unit, M112 ... image quality evaluation acquisition unit, M114 ... print setting determination unit

An ink jet that prints an image by forming dots (print pixels) by ejecting ink from the print head onto the paper while performing main scanning (also referred to as a pass) that moves the print head in the main scanning direction with respect to the paper. A printing device of the type is known. Further, by irradiating the lasers light to the photosensitive drum to form an electrostatic latent image on each main scanning line along the main scanning direction of the laser beam, the printing medium the toner adhered to the electrostatic latent image Laser printers that transfer are known. In these printing apparatuses, the image quality of the printed image may deteriorate due to deformation of the print medium .

The main advantage of the present invention is to provide a new technique to suppress deterioration of print quality due to the deformation of the printing medium.

Application Example 1 A print data generation device that generates print data for a print execution unit that forms print pixels on a print medium using a printing material,
The print execution unit
A print head having a plurality of nozzles for ejecting ink;
A main scanning unit that performs main scanning for moving the print head along the main scanning direction with respect to the print medium;
A sub-scanning unit that performs sub-scanning for moving the print medium along a sub-scanning direction that intersects the main scanning direction with respect to the print head, wherein a gap between the print medium and the nozzle is Executing the sub-scanning in a state where the print medium is deformed so as to change along the main-scanning direction;
A head drive unit that drives the print head during the main scan to form the print pixels on the print medium;
With
The print data generation device includes:
An image data acquisition unit for acquiring image data indicating an image composed of a plurality of pixels and including the multi-gradation values of each of the plurality of pixels, The number of gradations is greater than the type of print pixel formation state, the image data acquisition unit,
A halftone processing unit that performs a halftone process for converting the multi-gradation value of a pixel to be processed among the plurality of pixels into a print pixel value representing a formation state of the print pixel , the halftone process processing, using said multi-tone value of the target pixel, and a distribution error is processing for determining the print pixel value of the target pixel, the distribution error, the print pixel values already determined Of errors generated between the print pixel value in the determined pixel and the multi-gradation value in the determined pixel, which are distributed to the processing target pixel using an error matrix, and the error The matrix is a matrix that defines a distribution ratio of errors in peripheral pixels of the processing target pixel, the halftone processing unit ,
Equipped with a,
The halftone processing unit uses a plurality of the error matrix corresponding to said main scanning direction of the change in the gap between the nozzle and the print medium, that perform the halftone process, the print data Generator .

According to the arrangement, to suppress deterioration of print image quality attributable to deformation of the printing medium.

Claims (13)

A print control device for a print execution unit that forms print pixels on a print medium using a printing material,
Dot data that is a multi-gradation value that constitutes image data and has a number of gradations that is greater than the type of print pixel formation state, and that is composed of print pixel values that represent the print pixel formation state A halftone processing unit that executes a halftone process for converting to a halftone process, wherein the halftone process uses the multi-tone value of a pixel to be processed and a distribution error to use the print pixel value of the pixel to be processed. The distribution error is an error generated between the print pixel value in the determined pixel for which the print pixel value has been determined and the multi-gradation value in the determined pixel. The error is distributed to the processing target pixel using an error matrix, and the error matrix is a matrix that defines a distribution ratio of errors in peripheral pixels of the processing target pixel. And parts,
An error matrix acquisition unit that acquires a plurality of types of error matrices, wherein the plurality of types of error matrices include a first type of error matrix used when a pixel position shift occurs by a first amount, and the first type of error matrix. A second type of error matrix used when the pixel position shift occurs by a second amount different from the amount of the pixel, and the pixel position deviation includes the pixel position in the main scanning direction in the image data and the print medium. The error matrix acquisition unit, wherein the main scanning direction pixel position is a position along the main scanning direction of the peripheral pixels with respect to the processing target pixel; and
An evaluation image data generation unit that generates evaluation image print data representing a plurality of evaluation images corresponding to each of the plurality of types of error matrices by executing the halftone process using each of the plurality of types of error matrices. When,
A printing control apparatus. The print control apparatus according to claim 1,
The halftone processing unit is a first line matrix that defines the distribution ratio for a first peripheral pixel located in a first main scanning line to which the processing target pixel belongs, the error matrix, and the first matrix The halftone process is executed using the error matrix including a second line matrix that defines the distribution ratio for second peripheral pixels located on a second main scan line different from the one main scan line And
The first type error matrix is a reference error matrix used when the pixel positional deviation does not occur, and the second type error matrix is the main line error of the second line matrix of the reference error matrix. A printing control apparatus, which is a shift error matrix having the second line matrix shifted in the scanning direction. The print control apparatus according to claim 2,
The plurality of second-type error matrices include a first shift error matrix;
The second line matrix of the first shift error matrix is obtained by multiplying the distribution ratio of each pixel defined in the second line matrix of the reference error matrix by N (0 <N <1). A print control apparatus obtained by subtracting a multiplication distribution ratio from a distribution ratio of each pixel and further adding the multiplication distribution ratio to a distribution ratio of pixels adjacent to each pixel in a shift direction. The print control apparatus according to any one of claims 1 to 3,
The print execution unit includes a print head having a plurality of nozzles that eject ink, a main scanning unit that executes main scanning for moving the print head along the main scanning direction with respect to the print medium, and the print head A sub-scanning unit that performs sub-scanning that moves the print medium along a sub-scanning direction that intersects the main scanning direction, and drives the print head during the main scanning to print the print medium. A head driving unit for forming pixels,
The print execution unit is capable of executing bidirectional printing combining forward printing for forming the print pixels in the forward pass of the main scan and return pass printing for forming the print pixels in the return pass of the main scan,
Each of the plurality of types of error matrices is an error matrix set including an outward path error matrix and an inbound path error matrix,
The forward error matrix is used when the first main scanning line is printed by the forward printing and the second main scanning line is printed by the backward printing.
The return path error matrix is used when the first main scanning line is printed by the backward printing and the second main scanning line is printed by the forward printing.
The shift direction of the second line matrix of the forward path error matrix is opposite to the shift direction of the second line matrix of the backward path error matrix;
The evaluation image data generation unit generates the evaluation image print data representing a plurality of evaluation images by executing the halftone process using the error matrix set corresponding to each of the plurality of types of error matrix sets. A print control device. The print control apparatus according to claim 4,
The sub-scanning unit performs the sub-scanning in a state where the print medium is deformed so that a gap between the print medium and the nozzle changes along the main scanning direction;
The evaluation image data generation unit includes a plurality of first evaluation images printed in a first area of the print medium and a plurality of second evaluation images printed in a second area of the print medium. Generating the evaluation image print data representing,
The first region is a region where the gap is in a first range;
The printing control apparatus, wherein the second area is an area in which the gap is in a second range different from the first range. The print control apparatus according to claim 5,
The printing control apparatus, wherein the error matrix set corresponding to the first evaluation image and the error matrix set corresponding to the second evaluation image have different ranges of shift amounts of the second line matrix. The print control apparatus according to any one of claims 4 to 6,
The printing control apparatus, wherein a movement amount of the sub-scan performed between forward printing and backward printing for printing a plurality of evaluation images is the same as a printing pixel pitch in the sub-scanning direction. A print control apparatus according to any one of claims 1 to 7,
The print execution unit includes a print head having a plurality of nozzles that eject ink, a main scanning unit that executes main scanning for moving the print head along the main scanning direction with respect to the print medium, and the print head A sub-scanning unit that performs sub-scanning that moves the print medium along a sub-scanning direction that intersects the main scanning direction, and drives the print head during the main scanning to print the print medium. A head driving unit for forming pixels,
The plurality of nozzles of the print head are a plurality of the same color nozzles that eject the same color ink, and include the plurality of the same color nozzles arranged along the sub-scanning direction,
The print execution unit performs unidirectional printing using either one of the forward pass printing for forming the print pixels in the forward pass of the main scan and the return pass printing for forming the print pixels in the return pass of the main scan. Is feasible,
The evaluation image data generation unit is a print control device that generates the evaluation image print data representing a plurality of evaluation images printed by the one-way printing. The printing control apparatus according to claim 8,
The evaluation image data generation unit generates the evaluation image print data representing a plurality of evaluation images printed using the forward printing and a plurality of evaluation images printed using the backward printing. apparatus. The print control apparatus according to any one of claims 1 to 9, further comprising:
An evaluation acquisition unit that acquires an image quality evaluation for a print result printed using the evaluation image print data;
A print setting determination unit for determining a print setting to be used using the image quality evaluation;
A printing control apparatus. The print control apparatus according to claim 10,
The evaluation result acquisition unit includes a read data acquisition unit that acquires print result read data generated by reading a document printed using the evaluation image print data using an image reading device, and the print result read data. And an evaluation determining unit that determines the image quality evaluation. A print control apparatus according to any one of claims 1 to 11,
The evaluation image data generation unit generates the evaluation image print data so as to print a plurality of evaluation images in one main scan. A computer program for a print execution unit that forms print pixels on a print medium using a printing material,
Dot data that is a multi-gradation value that constitutes image data and has a number of gradations that is greater than the type of print pixel formation state, and that is composed of print pixel values that represent the print pixel formation state The halftone process is a first function for executing a halftone process for converting to the print pixel value of the processing target pixel by using the multi-gradation value of the processing target pixel and a distribution error. The distribution error is determined by using an error matrix among errors generated between the print pixel value and the multi-gradation value in the determined pixel for which the print pixel value has been determined. The first function, which is an error distributed to the processing target pixel, and the error matrix is a matrix defining a distribution ratio of errors with respect to peripheral pixels of the processing target pixel;
A second function for acquiring a plurality of types of error matrices, wherein the plurality of types of error matrices are a first type of error matrix used when a pixel position shift occurs by a first amount, and the first type of error matrix. A second type of error matrix used when the pixel position shift occurs by a second amount different from the amount of the pixel, and the pixel position deviation includes the pixel position in the main scanning direction in the image data and the print medium. The second function, wherein the main scanning direction pixel position is a position along the main scanning direction of the peripheral pixel with respect to the processing target pixel;
A third function for generating evaluation image print data representing a plurality of evaluation images corresponding to each of the plurality of types of error matrices by executing the halftone process using each of the plurality of types of error matrices; ,
A computer program that causes a computer to realize

Images